- Email: [email protected]
Endogenous catecholamines augment the inhibitory effect of opioids on luteinizing hormone secretion during the midluteal phase
Endogenous catecholamines augment the inhibitory effect of opioids on luteinizing hormone secretion during the midluteal phase Victor Y. Fujimoto, MD, Susan J. Spencer, MD, Jaron Rabinovici, MD, Shayne Plosker, MD. and Robert B. Jaffe. MD San Francisco, California OBJECTIVE: Our purpose was to test the hypothesis that endogenous catecholamines may interact with endogenous opioid peptides to influence gonadotropin secretion during the midluteal phase in normal women. STUDY DESIGN: Normal cycling women studied during the midluteal phase were randomized to one of four treatment groups: (1) a-methyl-para-tyrosine, (2) naloxone, (3) a-methyl-para-tyrosine and naloxone, and (4) control. Mean treatment luteinizing hormone levels were compared by analysis of variance. Pulse frequency, amplitude, and integrated area under the curve were assessed by CLUSTER analysis and compared by means of nonparametric analyses. RESULTS: Mean luteinizing hormone levels were significantly higher in the naloxone and a-methyl-para-tyrosine plus naloxone groups compared with a-methyl-para-tyrosine or placebo. Coadministration of a-methyl-para-tyrosine and naloxone caused a significant increase in large-burst luteinizing hormone pulses compared with the group receiving naloxone only. CONCLUSION: Endogenous catecholamines augment the inhibitory effect of opioids on luteinizing hormone secretion during the midluteal phase in normal cycling women. (AM J OBSTET GYNECOL 1993;169:1524-30.)
Key words: Endogenous catecholamines, naloxone, luteinizing hormone secretion, midluteal phase
The interaction between endogenous catecholamines and opioid peptides in the regulation of pituitary luteinizing hormone (LH) secretion in women remains poorly understood. Previous studies have identified synapses between dopaminergic, opioidergic, and gonadotropin-releasing hormone (GnRH) neurons in the hypothalamus of different species. I. 2 Kesner et at. 3 have shown that opiate administration to primates reduces basal and pulsatile gonadotropin secretion during the late follicular and midluteal phases of the menstrual cycle. The involvement of dopamine in gonadotropin From the Reproductive Endocrinology Center, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco. Supported in part by National Institutes of Health grant No. HD-1l729. These studies were carried out in the General Clinical Research Center, University of California, San Francisco, with funds provided by the Division of Research Resources, grant No. 5M01RR00079, United States Public Health Service. Presented in part at the Fortieth Annual Meeting of the Society for Gynecologic Investigation, Toronto, Ontario, Canada, March 31April 3, 1993. Reprint requests: Robert B. Jaffe, MD, Reproductive Endocrinology Center, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, San Francisco, CA 94143-0132. Copyright © 1993 by Mosby-Year Book, Inc. 0002-9378/93 $1.00 + .20 6/6/51007
1524
secretion in the human remains enigmatic. Some studies demonstrate the suppression of LH secretion by dopamine, whereas others do not.':" In our previous study, AMPT (ex-methyl-para-tyrosine), a catecholamine synthesis inhibitor, augmented LH pulse amplitude but not frequency during the early follicular phase of normally cycling women"; endogenous catecholamines inhibited LH secretion by damping pulse amplitude during the early follicular phase. Furthermore, it has been shown that the naloxone-induced rise in LH secretion can be reversed by concomitant intravenous administration of dopamine." Extending these observations, we developed an experimental protocol using AMPT' and naloxone, an opioid antagonist, to assess the potential interaction between endogenous catecholamines and opioid peptides in regulating LH secretion in the eumenorrheic state during the midluteal phase of normal cycling women.
Material and methods Subject selection. Sixteen healthy women of reproductive age (18 to 40 years old) of normal height and weight with regular ovulatory menstrual cycles (27 to 32 days), as determined by history, basal body temperature
Fujimoto et al. 1525
Volume 169, Number 6 Am J Obstet Gynecol
charting, and urinary LH testing or luteal phase progesterone measurements, were studied in the midluteal phase of 20 cycles between cycle days 21 and 23. Two women participated in two study cycles and one woman participated in three study cycles, with a minimum 3-month interval between any treatment cycle. In this double-blinded study repeat patients were placed into a different study arm in the subsequent treatment cycle by the General Clinical Research Center pharmacist, to minimize any inherent bias. All women had a normal screening medical history and physical examination. All participants had confirmed regular cycles by basal body temperature charting, early follicular phase serum follicle-stimulating hormone and LH values < 15 IU/L, and negative pregnancy tests, to exclude pregnancy or the perimenopausal state. Before participating in this study all subjects agreed to use nonhormonal contraception for the duration of the study. Other exclusion criteria included (1) elevated prolactin (PRL) levels, (2) hormonal medication use within 3 months of the proposed study cycle, (3) pregnancy or lactation within 6 months of the study, (4) narcotic abuse, or (5) abnormal medical findings on screening history, physical examination, or laboratory tests, including past history of extrapyramidal neurologic disease, depression, cardiovascular disease, psychiatric illness, or urolithiasis. Study protocol. Subjects were admitted to the General Clinical Research Center at 7 AM on one of menstrual cycle days 21 through 23. Confirmation of a midluteal-phase admission was made on the basis of basal body temperature charting and urinary LH kits. All participants reported normal menstrual cycle lengths during the study cycles. They were placed in a semirecumbent position, in which they remained for the duration of the study. An intravenous cannula, kept open with 0.9% saline solution and equipped with a four-way stopcock for frequent blood sampling, was placed in an antecubital or forearm vein. From 7:30 AM to 4 PM all subjects had blood drawn at lO-minute intervals; the study interval was 8.5 hours. Subjects received 500 mg of AMPT orally (Merck, Sharp & Dohme, West Point, Pa.) or placebo tablets at 8 and lOAM. This dose was selected on the basis of our previous study, 7 which demonstrated a significant rise in PRL and an absence of deleterious side effects. Naloxone (60 I-Lglkg/hr) or placebo (saline solution) was administered by infusion beginning at 8 AM and lasting until 4 PM. Subjects were not permitted to eat, drink, smoke, or sleep for the duration of the study. Sample preparation and hormone assay. All blood samples were assayed for LH, follicle-stimulating hormone, and PRL with radioimmunoassays previously described from these laboratories.v 10 Samples were centrifuged after 1 hour and the serum stored at - 20° C until assay. All samples from a single subject
were assayed in the same assay. Interassay coefficients of variation were < 15%, and intraassay coefficients of variation were 7% to 10% for both assays. The sensitivity of the LH assay (90% binding) was 1 IU/L. Values below the sensitivity of the assay were considered equal to the lower limit of sensitivity for any relevant calculations. Data analysis. Peripheral LH, follicle-stimulating hormone, and PRL levels were plotted as a function of time. Mean LH values for the different treatment groups were determined by averaging the LH values from all of the blood samples obtained after initiation of treatment from the five women within each group. Mean PRL values for each sampling time point were calculated for each of the four treatment groups. Pulsatile LH release was quantified objectively by CLUSTER analysis" with a defined coefficient of variation set at 10%. Area under the curve for each pulse, maximum pulse amplitude, and pulse frequency were calculated for each subject. Large-burst pulses were defined as any pulse with a maximum pulse slope > 0.65 IU/Umin. ' 2 Statistical analysis. Statistical significance of differences between the mean LH levels among the four treatment groups was determined by means of one-way analysis of variance with post hoc analysis using Fisher parametric least significant difference testing. The Kruskal-Wallis rank test was used to test for significant differences between large-burst LH pulse frequency, pulse amplitude, and area under the curve pulse between the study groups. When statistical significance was noted, the Mann-Whitney test was used to determine p values of differences between treatment groups. Nonparametric analyses were used when nonnormal distribution of a small sampling population with population variance differences between different treatment groups existed. Data are expressed as the mean ± SEM. Statistical significance was defined as a p value <0.05. Results
Endogenous catecholamine influence on PRL secretion. Similar to our previous report, AMPT had a dramatic and sustained effect on increasing PRL concentrations in all women who received the agent (Fig. 1). Basal PRL concentrations (mean ± SEM) were 18.7 ± 2.1 I-Lg/L and 13.0 ± 2.0 I-Lg/L in the AMPT only and AMPT-naloxone groups, respectively. Similarly, basal PRL levels were 18.2 ± 0.8 I-LgiL in the control group and 15.9 ± 0.8 I-Lg/L in the naloxoneonly group. After administration of AMPT, PRL concentrations rose immediately and remained elevated during the entire duration of sampling in both the AMPT only and AMPT-naloxone groups. The mean PRL concentration in the AMPT only group increased to 47.5 ± 1.1 I-Lg/L, which was not different from the
1526 Fujimoto et al.
December 1993 Am ] Obstet Gynecol
(8)
(b)
120
120
90
90
~
E
C>
a ..J
.s.'" ..J
60
60
a:
a:
l:I.
l:I.
30
30
0
0 6
0
Time (hours)
Time (hours)
(C)
120
90
~
~
..J
(d)
120
90
.s
8
0
8
.s
60
..J
60
a:
a:
l:I.
l:I.
30
30
0
0 6
0
0
TIme (hours)
Time (hours)
Fig. 1. Mean ± SEM serum PRL concentrations of five normal cycling women during following treatments: (a) placebo, (b) AMPT only, (c) naloxone only, and (d) AMPT and naloxone during 8-hour sampling period.
20
~
~
o
C
Placebo
r:ss
AMPT
IZI
Naloxone
•
AMPT/Naloxone
b
a 10
J:
...J
Fig. 2. Effect of AMPT and naloxone on mean ± SEM LH levels after initiation of treatment in all four treatment groups (n = 5). a, Value is significant compared with placebo (P < 0.05). b, Value is significant compared with placebo and AMPT groups I.P < 0.05). c, Value is significant compared with all three groups (p < 0.05).
mean concentration of 47.9 ± 2.4 I-Lg/L in the AMPTnaloxone group. Conversely, in the placebo and naloxone-only groups the mean PRL concentrations remained at 14.6 ± 0.6 and 12.9 ± 0.4 I-Lg/L, respectively. The only treatment side effect noted was mild sedation in some of the subjects receiving AMPT.
Endogenous catecholamine and opioid influence on LH secretion. As Fig. 2 demonstrates, the mean LH level after AMPT administration was significantly higher than after placebo (8.58 ± 0.29 vs 6.5 ± 0.29 lUlL, P < 0.05). Naloxone significantly increased mean LH levels (11.0 ± 0.3 Il.I/L) compared with placebo or AMPT (P < 0.05). Cotreatment with AMPT and naloxone increased the mean LH level to 16.22 ± 0.71 lUlL, significantly greater than all other treatment groups (P < 0.05). Fig. 3 illustrates individual LH pulse patterns in which LH values are plotted as a function of time for four representative subjects in each study group. Fig. 4 depicts LH pulse frequency patterns for the single patient who underwent three different treatment cycles. No pulses were detected with placebo, whereas two large-burst pulses occurred with naloxone alone and three large-burst pulses occurred with combined AMPT and naloxone. Fig. 5 shows the large-burst pulse frequency and pulse amplitude values (mean ± SEM) for the different treatment groups, as determined by the CLUSTER detection algorithm. AMPT administered alone increased the mean large-burst pulse frequency compared with the control group, but this was not statistically significant (1.0 ± 0.3 vs 0.4 ± 0.2 peaks per 8 hours, respectively) (P = 0.31). A greater increase in
Volume 169, Number 6 Am J Obstet Gynecol
Fujimoto et al. 1527
(a)
~
(b)
40
40
30
30
20
...:z:
10
10
0 ·0.67
0
4 Time (hours)
6
0+----._..------._,--,..----,,.--,---,_-,--,._.----.--. 0 4 6 8
-0.67
Time (hours)
(c)
~
i!
(d)
40
40
30
30
20
... :z:
10
10
0 -0.67
0+----.-.----.-...,----.,--,----.,--,----,,..---.----,,..--,---, 0
4 Time (hours)
6
·0.67
8
0 Time (hours)
Fig. 3. Individual LH pulse patterns during the 8-hour study period in four representative subjects during following treatments: (a) placebo, (b) AMPTonly, (c) naloxone only, (d) AMPT-naloxone. See text for details of experimental conditions.
the mean large-burst pulse frequency resulted from naloxone treatment (1.4 ± 0.4 peaks per 8 hours), which also failed to reach statistical significance compared with control (p = 0.08). Most strikingly, a dramatic increase in the number of large-burst LH pulses was seen in those patients treated concomitantly with AMPT and naloxone. The mean large-burst pulse frequency for the AMPT-naloxone-treated group was 3.0 ± 0.3 peaks per 8 hours, statistically greater during the study period than for the naloxone-only group (p = 0.02). When large-burst pulse amplitude was analyzed, neither AMPT nor naloxone had a significant effect on the pulse amplitude (p = 0.25). Confirming the notion that overall LH secretion per large burstpulse did not change, there was no difference in area under the curve pulse among any of the four groups tested (p = 0.82). Comment
AMPT and naloxone given together markedly increased the frequency of large-burst pulses occurring during the 8-hour sampling period compared with that
of the naloxone-only group. Taken together with our findings of increased mean posttreatment LH levels in both naloxone- and AMPT-naloxone-treated groups compared with placebo and AMPT only, these data strongly suggest that endogenous catecholamines augment the inhibitory effect of opioid peptides on LH secretion during the midluteal phase of the human menstrual cycle, a period of high estrogen and progesterone secretion. The inhibition of LH secretion by endogenous opioids demonstrated in previous work 1 3. I. was confirmed in this study. Although statistical significance was not achieved, our data suggest that this inhibition may occur through a decrease in pulse frequency, but not amplitude, of the large-burst pulses characteristic of the luteal phase. AMPT administered alone increased mean posttreatment LH levels and LH pulse frequency compared with placebo, but this latter difference did not reach statistical significance. A longer sampling period may be required to demonstrate a significant effect of AMPT on LH pulsatility. Although there are reports of AMPT administered to normal cycling women during the follicular phase," 15, 16
1528 Fujimoto at al.
December 1993 Am J Obstet Gynecol
(a)
(a)
4
40
D
Placebo
~ AMPT 3
:c e
30
~ Naloxone •
*
AMPT/Naloxone
~ c
::r
! ~ II. ]I
:;
;. 20
3
2
t
10
0+---.-..,.---.-..,.---.-..,.---.-..,.---.-..--.-..---. -4.67
0
2
4 Time (hours)
6
0
8
(b)
(b) 40
20
::J :5 e
30
Ql
'tl
a Q.
E
ce
10
]I ~
ll.
10
0+---.-..,.---.-..--.-..--.-..--.-..--.-..---. -4.67
0
2
4 Time (hours)
6
8
Fig. 5. Effect of AMPT and naloxone on (a) mean ±SEM large-burst LH pulse frequency and (b) mean ± SEM largeburst LH pulse amplitude during midluteal phase of normal cycling women (n = 5 per treatment group). Asterisk, Value is significant compared with group receiving naloxone only
(c)
80
(P
=
0.02).
60
~
;. 40
:5 20
2
4
6
8
Time (hours)
Fig. 4. LH pulse patterns in single patient undergoing (a) placebo, (b) naloxone only, (c) AMPT-naloxone. See text for details of experimental conditions.
it has never been administered during the luteal phase of the normal menstrual cycle, a time of relatively high levels of endogenous estrogens and progesterone. Our study design addresses the interaction of endogenous catecholamines and opiates on the hypothalamic-pituitary-ovarian axis by administering both AMPT and naloxone. This study demonstrates both endogenous
catecholaminergic and opioidergic effects on LH secretion during the midluteal phase. Similar to the findings of Plosker et a1.,7 in our study administration of AMPT resulted in a marked and prolonged elevation of serum PRL levels, irrespective of naloxone status. Our study did not confirm the naloxone-induced serum PRL rise seen during the midluteal phase reported by Cetel et a1. 17 It is likely that AMPT augments the naloxone response by decreasing endogenous hypothalamic dopamine production. The marked increase in PRL secretion occurs through the inhibition of tuberoinfundibular dopamine production by AMPT. Previous studies in which dopaminergic regulation of gonadotropin secretion was assessed in women indicate inhibition by dopamine.": 18 When metaclopramide, a dopamine receptor antagonist, was given intravenously to women in the midluteal phase, an increase in serum LH levels was noted." In contrast, treatment with the a-adrenergic receptor antagonists phenoxybenzamine and phentol-
Fujimoto at al. 1529
Volume 169, Number 6 AmJ Obstet Gynecol
amine resulted in the inhibition of pulsatile LH release in ovariectomized monkeys, indicating a stimulatory action of norepinephrine in the primate." There appears to be no role for epinephrine in the regulation of pulsatile LH release, because specific inhibition of epinephrine synthesis does not affect pulsatile release." It is possible that the effect of AMPT in augmenting the naloxone-induced increase of LH pulsatility is a result of elevated PRL levels. However, we did not observe a significant increase in LH pulsatility in the group given AMPT alone, in which there also were elevated levels of PRL. The demonstration of neurotransmitter interaction in regulating LH secretion raises important questions regarding the morphologic neuronal network regulating the pulse generator. The hierachy of hypothalmic opiate and catecholamine effects on gonadotropin secretion remains unclear in spite of previous studies in different animal models. In rat medial basal hypothalamic slices, naloxone blocked the inhibitory effect of the selective K-agonist U 50,488 and the selective K-agonist (D-Ala 2 , N-Me-Phe" Cly-oll-enkephalin on tritiated dopamine release." The addition of naloxone stimulated dopamine, norepinephrine, epinephrine, and GnRH release in perifused rat preoptic areamedial basal hypothalamic fragments." These studies suggest direct regulation of hypothalamic dopamine release by opiates in the rat. However, Rasmussen et a1. 23 reported that dopamine stimulated l3-endorphin release in vitro from adult human hypothalami. Thus it has been theorized that dopaminergic activity may alter GnRH secretion by increasing opioid tone." Direct inhibitory synapses exist between opioid and GnRH neurons in the juvenile monkey at the level of the arcuate nucleus-infundibular lip! In the rhesus monkey median eminence, direct dopaminergic-GnRH contacts without synaptic specializations appear to exist. 25 Taken together, the body of evidence indicates direct inhibition of GnRH neuronal pulsatility by both opiates and dopamine at different anatomic levels in the primate hypothalamus. In the primate pituitary, portal blood l3-endorphin levels are steroid dependent and highest during the midluteal phase, when levels of both estrogen and progesterone are high.?" Progesterone increases l3-endorphin secretion into portal blood of ovariectomized monkeys treated with estrogen." Immunocytochemical colocalization of hypothalamic progesterone receptors and tyrosine hydroxylase was found in the dopaminergic neurons of steroid-treated macaques." A previous study in women suggested a functional association between dopaminergic, opioidergic, and GnRH neuronal systems. Intravenous dopamine infused at 2 ILg/kg/min with naloxone completely reversed the naloxone-induced rise in LH pulsatility seen with naloxone admin-
istration alone in the midluteal phase of the menstrual
cycle." When the dopamine precursor L-dopa-C was administered to hyperandrogenic anovulatory women, there was an increased LH response to opiate receptor blockade." Our findings confirm the inhibitory effects of catecholamines and opiates on LH secretion seen in these studies. Although our data suggest the steroiddependent nature of hypothalamic dopaminergicopioidergic regulation, chronic administration of veralipride, a dopaminergic-receptor antagonist, to postmenopausal women restored the stimulatory effect of naloxone on LH secretion." In summary, these findings demonstrate that both endogenous opioid and catecholamine regulation of the hypothalamic-pituitary-gonadal axis exists during the midluteal phase of the menstrual cycle. The observation that AMPT and naloxone given together increase pulse frequency dramatically suggests independent effects at the level of the hypothalamic pulse generator. Together with morphologic observations in the primate, these data support direct control of GnRH pulsatility by both dopamine and opiates. We thank Carmelita Aquirre, Steven Zippin, Deborah Downey, and Cindy Voytek for excellent assistance. REFERENCES 1. Morel G, Pelletier G. Endorphinic neurons are contacting the tuberoinfundibular dopaminergic neurons in the rat brain. Peptides 1986;7: I 197-9. 2. Thind K, Goldsmith P. Infundibular gonadotropin-releasing hormone neurons are inhibited by direct opioid and autoregulatory synapses in juvenile monkeys. Neuroendocrinology 1988;47:203-16. 3. Kesner J, Kaufman J, Wilson R, Kuroda G, Knobil E. The effect of morphine on the electrophysiological activity of the hypothalamic luteinizing hormone-releasing hormone pulse generator in the rhesus monkey. Neuroendocrinology 1986;43:686-8. 4. Judd S, Rakoff J, Yen S. Inhibition of gonadotropin and prolactin release by dopamine: effect of endogenous estradiol levels. J Clin Endocrinol Metab 1978;47:494-8. 5. Lachelin G, LeBlanc H, Yen S. The inhibitory effect of dopamine agonists on LH release in women. J Clin Endocrinol Metab 1977;44:728-32. 6. Martin M, Monroe S, Weiner R, Jaffe R. Low-dose dopamine infusions do not ablate luteinizing hormone pulses in women. AM J OBSTET GYNECOL 1988;159:898-903. 7. Plosker S, Rabinovici J, Jaffe R. Inhibition of endogenous catecholamine synthesis augments early follicular phase luteinizing hormone secretion. J Clin Endocrinol Metab 1991;73:549-54. 8. Ropert ], Quigley M, Yen S. The dopaminergic inhibition of LH secretion during the menstrual cycle. Life Sci 1984;34:2067-73. 9. Engelman K, Horwitz D, Jequier E, Sjoerdsma A. Biochemical and pharmacologic effects of o-methyltyrosine in man. J Clin Invest 1968;47:577-94. 10. Martin M, Weiner R, Monroe S, Roberts ], Licko V,Jaffe R. Prolactin-secreting adenomas in women. VII. Dopamine regulation of prolactin secretion. J Clin Endocrinol Metab 1984;59:485-90. II. Veldhuis J, Johnson M. Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol 1986;250:E486-93.
1530 Fujimoto at al.
12. Sollenberger MJ, Carlsen EC, Booth RAJr, Johnson ML, Veldhuis JD, Evans WS. Nature of gonadotropin-releasing hormone self-priming of luteinizing hormone secretion during the normal menstrual cycle. AM J OBSTET GYNECOL 1990; 163:1529-34. 13. Quigley M, Yen S. The role of endogenous opiates on LH secretion during the menstrual cycle. J Clin Endocrinol Metab 1980;51:179-81. 14. Ropert J, Quigley M, Yen S. Endogenous opiates modulate pulsatile luteinizing hormone release in humans. J Clin Endocrinol Metab 1981;52:583-5. 15. Nicoletti I, Filipponi P, Sfrappini M. Catecholamines and pituitary function. I. Effects of catecholamine synthesis inhibition and subsequent catecholamine infusion on gonadotropin and prolactin serum levels in normal cycling women and in women with hyperprolactinemic amenorrhea. Horm Res 1984;19:158-70. 16. Nicoletti I, Filipponi P, Fedeli L, Angeletti G, Santeusanio F, Brunetti P. Effect of methyl-p-tyrosine administration on gonadotropin and prolactin release in normal-cycling, postmenopausal, and hyperprolactinemic women. Acta Eur Fertil 1981;12:271-3. 17. Cetel N, Quigley M, Yen S. Naloxone-induced prolactin secretion in women: evidence against a direct prolactin stimulatory effect of endogenous opioids. J Clin Endocrinol Metab 1985;60:191-6. 18. Judd S, Rakoff J, Yen S. The effects of ovariectomy and estrogen treatment on the dopamine inhibition of gonadotropin and prolactin release. J Clin Endocrinol Metab 1979;49: 182-4. 19. Bhattacharya A, Dierschke D, Yamaji J, Knobil E. The pharmacologic blockade of the circhoral mode of LH secretion in the ovariectomized rhesus monkey. Endocrinology 1972;90:778-86. 20. Crowley W, Terry L, Johnson M. Evidence for the involvement of central epinephrine systems in the regulation of luteinizing hormone, prolactin, and growth hormone release in female rats. Endocrinology 1982; 110: 1102-7. 21. Heijna M, Padt M, Hogenboom F, Schoffelmeer A, Mulder
December 1993 Am J Obstet Gynecol
A. Opioid-receptor-mediated inhibition of [3H]dopamine
22.
23.
24.
25.
26. 27. 28.
29. 30.
but not [3H]noradrenaline release from rat mediobasal hypothalamus slices. Neuroendocrinology 1991 ;54: 11826. Leadem C, Crowley W, Simpkins J, Kalra S. Effects of naloxone on catecholamine and LHRH release from the perifused hypothalamus of the steroid-primed rat. Neuroendocrinology 1985;40:497-500. Rasmussen D, Liu J, Wolf P, Yen S. Neurosecretion of human hypothalamic immunoreactive l3-endorphin: in vitro regulation by dopamine. Neuroendocrinology 1987; 45:197-200. Sarkar D, Yen S. Hyperprolactinemia decreases the luteinizing hormone-releasing hormone concentration in pituitary portal plasma: a possible role for l3-endorphin as a mediator. Endocrinology 1985;116:2080-4. Goldsmith P, Thind K. Modulation of GnRH pulse generator activity. In: Genazzani AR, Montemagno D, Nappi C, Petraglia F, eds. The brain and female reproductive function. Lancaster, England: Parthenon, 1987:21-33. Wehrenberg W, Wardlaw S, Frantz A, Ferin M. I3-Endorphin in hypophyseal portal blood: variations throughout the menstrual cycle. Endocrinology 1982; 111:879-81. Wardlaw S, Wehrenburg W, Ferin M, Antunes J, Frantz A. Effect of sex steroids on l3-endorphin in hypophyseal portal blood. J Clin Endocrinol Metab 1982;55:877-81. Kohama S, Freesh F, Bethea C. Immunocytochemical colocalization of hypothalamic progestin receptors and tyrosine hydroxylase in steroid-treated monkeys. Endocrinology 1992;131:509-17. Barnes R, Lobo R. Central opioid activity in polycystic ovary syndrome with and without dopaminergic modulation. J Clin Endocrinol Metab 1985;61:779-82. Melis G, Cagnacci A, Gambacciani M, Paoletti A, Moggi L, Fioretti P. Restoration ofluteinizing hormone response to naloxone in postmenopausal women by chronic administration of the anti-dopaminergic drug veralipride. J Clin Endocrinol Metab 1988;66:964-7.
Key words: Endogenous catecholamines, naloxone, luteinizing hormone secretion, midluteal phase
The interaction between endogenous catecholamines and opioid peptides in the regulation of pituitary luteinizing hormone (LH) secretion in women remains poorly understood. Previous studies have identified synapses between dopaminergic, opioidergic, and gonadotropin-releasing hormone (GnRH) neurons in the hypothalamus of different species. I. 2 Kesner et at. 3 have shown that opiate administration to primates reduces basal and pulsatile gonadotropin secretion during the late follicular and midluteal phases of the menstrual cycle. The involvement of dopamine in gonadotropin From the Reproductive Endocrinology Center, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco. Supported in part by National Institutes of Health grant No. HD-1l729. These studies were carried out in the General Clinical Research Center, University of California, San Francisco, with funds provided by the Division of Research Resources, grant No. 5M01RR00079, United States Public Health Service. Presented in part at the Fortieth Annual Meeting of the Society for Gynecologic Investigation, Toronto, Ontario, Canada, March 31April 3, 1993. Reprint requests: Robert B. Jaffe, MD, Reproductive Endocrinology Center, Department of Obstetrics, Gynecology and Reproductive Sciences, University of California, San Francisco, San Francisco, CA 94143-0132. Copyright © 1993 by Mosby-Year Book, Inc. 0002-9378/93 $1.00 + .20 6/6/51007
1524
secretion in the human remains enigmatic. Some studies demonstrate the suppression of LH secretion by dopamine, whereas others do not.':" In our previous study, AMPT (ex-methyl-para-tyrosine), a catecholamine synthesis inhibitor, augmented LH pulse amplitude but not frequency during the early follicular phase of normally cycling women"; endogenous catecholamines inhibited LH secretion by damping pulse amplitude during the early follicular phase. Furthermore, it has been shown that the naloxone-induced rise in LH secretion can be reversed by concomitant intravenous administration of dopamine." Extending these observations, we developed an experimental protocol using AMPT' and naloxone, an opioid antagonist, to assess the potential interaction between endogenous catecholamines and opioid peptides in regulating LH secretion in the eumenorrheic state during the midluteal phase of normal cycling women.
Material and methods Subject selection. Sixteen healthy women of reproductive age (18 to 40 years old) of normal height and weight with regular ovulatory menstrual cycles (27 to 32 days), as determined by history, basal body temperature
Fujimoto et al. 1525
Volume 169, Number 6 Am J Obstet Gynecol
charting, and urinary LH testing or luteal phase progesterone measurements, were studied in the midluteal phase of 20 cycles between cycle days 21 and 23. Two women participated in two study cycles and one woman participated in three study cycles, with a minimum 3-month interval between any treatment cycle. In this double-blinded study repeat patients were placed into a different study arm in the subsequent treatment cycle by the General Clinical Research Center pharmacist, to minimize any inherent bias. All women had a normal screening medical history and physical examination. All participants had confirmed regular cycles by basal body temperature charting, early follicular phase serum follicle-stimulating hormone and LH values < 15 IU/L, and negative pregnancy tests, to exclude pregnancy or the perimenopausal state. Before participating in this study all subjects agreed to use nonhormonal contraception for the duration of the study. Other exclusion criteria included (1) elevated prolactin (PRL) levels, (2) hormonal medication use within 3 months of the proposed study cycle, (3) pregnancy or lactation within 6 months of the study, (4) narcotic abuse, or (5) abnormal medical findings on screening history, physical examination, or laboratory tests, including past history of extrapyramidal neurologic disease, depression, cardiovascular disease, psychiatric illness, or urolithiasis. Study protocol. Subjects were admitted to the General Clinical Research Center at 7 AM on one of menstrual cycle days 21 through 23. Confirmation of a midluteal-phase admission was made on the basis of basal body temperature charting and urinary LH kits. All participants reported normal menstrual cycle lengths during the study cycles. They were placed in a semirecumbent position, in which they remained for the duration of the study. An intravenous cannula, kept open with 0.9% saline solution and equipped with a four-way stopcock for frequent blood sampling, was placed in an antecubital or forearm vein. From 7:30 AM to 4 PM all subjects had blood drawn at lO-minute intervals; the study interval was 8.5 hours. Subjects received 500 mg of AMPT orally (Merck, Sharp & Dohme, West Point, Pa.) or placebo tablets at 8 and lOAM. This dose was selected on the basis of our previous study, 7 which demonstrated a significant rise in PRL and an absence of deleterious side effects. Naloxone (60 I-Lglkg/hr) or placebo (saline solution) was administered by infusion beginning at 8 AM and lasting until 4 PM. Subjects were not permitted to eat, drink, smoke, or sleep for the duration of the study. Sample preparation and hormone assay. All blood samples were assayed for LH, follicle-stimulating hormone, and PRL with radioimmunoassays previously described from these laboratories.v 10 Samples were centrifuged after 1 hour and the serum stored at - 20° C until assay. All samples from a single subject
were assayed in the same assay. Interassay coefficients of variation were < 15%, and intraassay coefficients of variation were 7% to 10% for both assays. The sensitivity of the LH assay (90% binding) was 1 IU/L. Values below the sensitivity of the assay were considered equal to the lower limit of sensitivity for any relevant calculations. Data analysis. Peripheral LH, follicle-stimulating hormone, and PRL levels were plotted as a function of time. Mean LH values for the different treatment groups were determined by averaging the LH values from all of the blood samples obtained after initiation of treatment from the five women within each group. Mean PRL values for each sampling time point were calculated for each of the four treatment groups. Pulsatile LH release was quantified objectively by CLUSTER analysis" with a defined coefficient of variation set at 10%. Area under the curve for each pulse, maximum pulse amplitude, and pulse frequency were calculated for each subject. Large-burst pulses were defined as any pulse with a maximum pulse slope > 0.65 IU/Umin. ' 2 Statistical analysis. Statistical significance of differences between the mean LH levels among the four treatment groups was determined by means of one-way analysis of variance with post hoc analysis using Fisher parametric least significant difference testing. The Kruskal-Wallis rank test was used to test for significant differences between large-burst LH pulse frequency, pulse amplitude, and area under the curve pulse between the study groups. When statistical significance was noted, the Mann-Whitney test was used to determine p values of differences between treatment groups. Nonparametric analyses were used when nonnormal distribution of a small sampling population with population variance differences between different treatment groups existed. Data are expressed as the mean ± SEM. Statistical significance was defined as a p value <0.05. Results
Endogenous catecholamine influence on PRL secretion. Similar to our previous report, AMPT had a dramatic and sustained effect on increasing PRL concentrations in all women who received the agent (Fig. 1). Basal PRL concentrations (mean ± SEM) were 18.7 ± 2.1 I-Lg/L and 13.0 ± 2.0 I-Lg/L in the AMPT only and AMPT-naloxone groups, respectively. Similarly, basal PRL levels were 18.2 ± 0.8 I-LgiL in the control group and 15.9 ± 0.8 I-Lg/L in the naloxoneonly group. After administration of AMPT, PRL concentrations rose immediately and remained elevated during the entire duration of sampling in both the AMPT only and AMPT-naloxone groups. The mean PRL concentration in the AMPT only group increased to 47.5 ± 1.1 I-Lg/L, which was not different from the
1526 Fujimoto et al.
December 1993 Am ] Obstet Gynecol
(8)
(b)
120
120
90
90
~
E
C>
a ..J
.s.'" ..J
60
60
a:
a:
l:I.
l:I.
30
30
0
0 6
0
Time (hours)
Time (hours)
(C)
120
90
~
~
..J
(d)
120
90
.s
8
0
8
.s
60
..J
60
a:
a:
l:I.
l:I.
30
30
0
0 6
0
0
TIme (hours)
Time (hours)
Fig. 1. Mean ± SEM serum PRL concentrations of five normal cycling women during following treatments: (a) placebo, (b) AMPT only, (c) naloxone only, and (d) AMPT and naloxone during 8-hour sampling period.
20
~
~
o
C
Placebo
r:ss
AMPT
IZI
Naloxone
•
AMPT/Naloxone
b
a 10
J:
...J
Fig. 2. Effect of AMPT and naloxone on mean ± SEM LH levels after initiation of treatment in all four treatment groups (n = 5). a, Value is significant compared with placebo (P < 0.05). b, Value is significant compared with placebo and AMPT groups I.P < 0.05). c, Value is significant compared with all three groups (p < 0.05).
mean concentration of 47.9 ± 2.4 I-Lg/L in the AMPTnaloxone group. Conversely, in the placebo and naloxone-only groups the mean PRL concentrations remained at 14.6 ± 0.6 and 12.9 ± 0.4 I-Lg/L, respectively. The only treatment side effect noted was mild sedation in some of the subjects receiving AMPT.
Endogenous catecholamine and opioid influence on LH secretion. As Fig. 2 demonstrates, the mean LH level after AMPT administration was significantly higher than after placebo (8.58 ± 0.29 vs 6.5 ± 0.29 lUlL, P < 0.05). Naloxone significantly increased mean LH levels (11.0 ± 0.3 Il.I/L) compared with placebo or AMPT (P < 0.05). Cotreatment with AMPT and naloxone increased the mean LH level to 16.22 ± 0.71 lUlL, significantly greater than all other treatment groups (P < 0.05). Fig. 3 illustrates individual LH pulse patterns in which LH values are plotted as a function of time for four representative subjects in each study group. Fig. 4 depicts LH pulse frequency patterns for the single patient who underwent three different treatment cycles. No pulses were detected with placebo, whereas two large-burst pulses occurred with naloxone alone and three large-burst pulses occurred with combined AMPT and naloxone. Fig. 5 shows the large-burst pulse frequency and pulse amplitude values (mean ± SEM) for the different treatment groups, as determined by the CLUSTER detection algorithm. AMPT administered alone increased the mean large-burst pulse frequency compared with the control group, but this was not statistically significant (1.0 ± 0.3 vs 0.4 ± 0.2 peaks per 8 hours, respectively) (P = 0.31). A greater increase in
Volume 169, Number 6 Am J Obstet Gynecol
Fujimoto et al. 1527
(a)
~
(b)
40
40
30
30
20
...:z:
10
10
0 ·0.67
0
4 Time (hours)
6
0+----._..------._,--,..----,,.--,---,_-,--,._.----.--. 0 4 6 8
-0.67
Time (hours)
(c)
~
i!
(d)
40
40
30
30
20
... :z:
10
10
0 -0.67
0+----.-.----.-...,----.,--,----.,--,----,,..---.----,,..--,---, 0
4 Time (hours)
6
·0.67
8
0 Time (hours)
Fig. 3. Individual LH pulse patterns during the 8-hour study period in four representative subjects during following treatments: (a) placebo, (b) AMPTonly, (c) naloxone only, (d) AMPT-naloxone. See text for details of experimental conditions.
the mean large-burst pulse frequency resulted from naloxone treatment (1.4 ± 0.4 peaks per 8 hours), which also failed to reach statistical significance compared with control (p = 0.08). Most strikingly, a dramatic increase in the number of large-burst LH pulses was seen in those patients treated concomitantly with AMPT and naloxone. The mean large-burst pulse frequency for the AMPT-naloxone-treated group was 3.0 ± 0.3 peaks per 8 hours, statistically greater during the study period than for the naloxone-only group (p = 0.02). When large-burst pulse amplitude was analyzed, neither AMPT nor naloxone had a significant effect on the pulse amplitude (p = 0.25). Confirming the notion that overall LH secretion per large burstpulse did not change, there was no difference in area under the curve pulse among any of the four groups tested (p = 0.82). Comment
AMPT and naloxone given together markedly increased the frequency of large-burst pulses occurring during the 8-hour sampling period compared with that
of the naloxone-only group. Taken together with our findings of increased mean posttreatment LH levels in both naloxone- and AMPT-naloxone-treated groups compared with placebo and AMPT only, these data strongly suggest that endogenous catecholamines augment the inhibitory effect of opioid peptides on LH secretion during the midluteal phase of the human menstrual cycle, a period of high estrogen and progesterone secretion. The inhibition of LH secretion by endogenous opioids demonstrated in previous work 1 3. I. was confirmed in this study. Although statistical significance was not achieved, our data suggest that this inhibition may occur through a decrease in pulse frequency, but not amplitude, of the large-burst pulses characteristic of the luteal phase. AMPT administered alone increased mean posttreatment LH levels and LH pulse frequency compared with placebo, but this latter difference did not reach statistical significance. A longer sampling period may be required to demonstrate a significant effect of AMPT on LH pulsatility. Although there are reports of AMPT administered to normal cycling women during the follicular phase," 15, 16
1528 Fujimoto at al.
December 1993 Am J Obstet Gynecol
(a)
(a)
4
40
D
Placebo
~ AMPT 3
:c e
30
~ Naloxone •
*
AMPT/Naloxone
~ c
::r
! ~ II. ]I
:;
;. 20
3
2
t
10
0+---.-..,.---.-..,.---.-..,.---.-..,.---.-..--.-..---. -4.67
0
2
4 Time (hours)
6
0
8
(b)
(b) 40
20
::J :5 e
30
Ql
'tl
a Q.
E
ce
10
]I ~
ll.
10
0+---.-..,.---.-..--.-..--.-..--.-..--.-..---. -4.67
0
2
4 Time (hours)
6
8
Fig. 5. Effect of AMPT and naloxone on (a) mean ±SEM large-burst LH pulse frequency and (b) mean ± SEM largeburst LH pulse amplitude during midluteal phase of normal cycling women (n = 5 per treatment group). Asterisk, Value is significant compared with group receiving naloxone only
(c)
80
(P
=
0.02).
60
~
;. 40
:5 20
2
4
6
8
Time (hours)
Fig. 4. LH pulse patterns in single patient undergoing (a) placebo, (b) naloxone only, (c) AMPT-naloxone. See text for details of experimental conditions.
it has never been administered during the luteal phase of the normal menstrual cycle, a time of relatively high levels of endogenous estrogens and progesterone. Our study design addresses the interaction of endogenous catecholamines and opiates on the hypothalamic-pituitary-ovarian axis by administering both AMPT and naloxone. This study demonstrates both endogenous
catecholaminergic and opioidergic effects on LH secretion during the midluteal phase. Similar to the findings of Plosker et a1.,7 in our study administration of AMPT resulted in a marked and prolonged elevation of serum PRL levels, irrespective of naloxone status. Our study did not confirm the naloxone-induced serum PRL rise seen during the midluteal phase reported by Cetel et a1. 17 It is likely that AMPT augments the naloxone response by decreasing endogenous hypothalamic dopamine production. The marked increase in PRL secretion occurs through the inhibition of tuberoinfundibular dopamine production by AMPT. Previous studies in which dopaminergic regulation of gonadotropin secretion was assessed in women indicate inhibition by dopamine.": 18 When metaclopramide, a dopamine receptor antagonist, was given intravenously to women in the midluteal phase, an increase in serum LH levels was noted." In contrast, treatment with the a-adrenergic receptor antagonists phenoxybenzamine and phentol-
Fujimoto at al. 1529
Volume 169, Number 6 AmJ Obstet Gynecol
amine resulted in the inhibition of pulsatile LH release in ovariectomized monkeys, indicating a stimulatory action of norepinephrine in the primate." There appears to be no role for epinephrine in the regulation of pulsatile LH release, because specific inhibition of epinephrine synthesis does not affect pulsatile release." It is possible that the effect of AMPT in augmenting the naloxone-induced increase of LH pulsatility is a result of elevated PRL levels. However, we did not observe a significant increase in LH pulsatility in the group given AMPT alone, in which there also were elevated levels of PRL. The demonstration of neurotransmitter interaction in regulating LH secretion raises important questions regarding the morphologic neuronal network regulating the pulse generator. The hierachy of hypothalmic opiate and catecholamine effects on gonadotropin secretion remains unclear in spite of previous studies in different animal models. In rat medial basal hypothalamic slices, naloxone blocked the inhibitory effect of the selective K-agonist U 50,488 and the selective K-agonist (D-Ala 2 , N-Me-Phe" Cly-oll-enkephalin on tritiated dopamine release." The addition of naloxone stimulated dopamine, norepinephrine, epinephrine, and GnRH release in perifused rat preoptic areamedial basal hypothalamic fragments." These studies suggest direct regulation of hypothalamic dopamine release by opiates in the rat. However, Rasmussen et a1. 23 reported that dopamine stimulated l3-endorphin release in vitro from adult human hypothalami. Thus it has been theorized that dopaminergic activity may alter GnRH secretion by increasing opioid tone." Direct inhibitory synapses exist between opioid and GnRH neurons in the juvenile monkey at the level of the arcuate nucleus-infundibular lip! In the rhesus monkey median eminence, direct dopaminergic-GnRH contacts without synaptic specializations appear to exist. 25 Taken together, the body of evidence indicates direct inhibition of GnRH neuronal pulsatility by both opiates and dopamine at different anatomic levels in the primate hypothalamus. In the primate pituitary, portal blood l3-endorphin levels are steroid dependent and highest during the midluteal phase, when levels of both estrogen and progesterone are high.?" Progesterone increases l3-endorphin secretion into portal blood of ovariectomized monkeys treated with estrogen." Immunocytochemical colocalization of hypothalamic progesterone receptors and tyrosine hydroxylase was found in the dopaminergic neurons of steroid-treated macaques." A previous study in women suggested a functional association between dopaminergic, opioidergic, and GnRH neuronal systems. Intravenous dopamine infused at 2 ILg/kg/min with naloxone completely reversed the naloxone-induced rise in LH pulsatility seen with naloxone admin-
istration alone in the midluteal phase of the menstrual
cycle." When the dopamine precursor L-dopa-C was administered to hyperandrogenic anovulatory women, there was an increased LH response to opiate receptor blockade." Our findings confirm the inhibitory effects of catecholamines and opiates on LH secretion seen in these studies. Although our data suggest the steroiddependent nature of hypothalamic dopaminergicopioidergic regulation, chronic administration of veralipride, a dopaminergic-receptor antagonist, to postmenopausal women restored the stimulatory effect of naloxone on LH secretion." In summary, these findings demonstrate that both endogenous opioid and catecholamine regulation of the hypothalamic-pituitary-gonadal axis exists during the midluteal phase of the menstrual cycle. The observation that AMPT and naloxone given together increase pulse frequency dramatically suggests independent effects at the level of the hypothalamic pulse generator. Together with morphologic observations in the primate, these data support direct control of GnRH pulsatility by both dopamine and opiates. We thank Carmelita Aquirre, Steven Zippin, Deborah Downey, and Cindy Voytek for excellent assistance. REFERENCES 1. Morel G, Pelletier G. Endorphinic neurons are contacting the tuberoinfundibular dopaminergic neurons in the rat brain. Peptides 1986;7: I 197-9. 2. Thind K, Goldsmith P. Infundibular gonadotropin-releasing hormone neurons are inhibited by direct opioid and autoregulatory synapses in juvenile monkeys. Neuroendocrinology 1988;47:203-16. 3. Kesner J, Kaufman J, Wilson R, Kuroda G, Knobil E. The effect of morphine on the electrophysiological activity of the hypothalamic luteinizing hormone-releasing hormone pulse generator in the rhesus monkey. Neuroendocrinology 1986;43:686-8. 4. Judd S, Rakoff J, Yen S. Inhibition of gonadotropin and prolactin release by dopamine: effect of endogenous estradiol levels. J Clin Endocrinol Metab 1978;47:494-8. 5. Lachelin G, LeBlanc H, Yen S. The inhibitory effect of dopamine agonists on LH release in women. J Clin Endocrinol Metab 1977;44:728-32. 6. Martin M, Monroe S, Weiner R, Jaffe R. Low-dose dopamine infusions do not ablate luteinizing hormone pulses in women. AM J OBSTET GYNECOL 1988;159:898-903. 7. Plosker S, Rabinovici J, Jaffe R. Inhibition of endogenous catecholamine synthesis augments early follicular phase luteinizing hormone secretion. J Clin Endocrinol Metab 1991;73:549-54. 8. Ropert ], Quigley M, Yen S. The dopaminergic inhibition of LH secretion during the menstrual cycle. Life Sci 1984;34:2067-73. 9. Engelman K, Horwitz D, Jequier E, Sjoerdsma A. Biochemical and pharmacologic effects of o-methyltyrosine in man. J Clin Invest 1968;47:577-94. 10. Martin M, Weiner R, Monroe S, Roberts ], Licko V,Jaffe R. Prolactin-secreting adenomas in women. VII. Dopamine regulation of prolactin secretion. J Clin Endocrinol Metab 1984;59:485-90. II. Veldhuis J, Johnson M. Cluster analysis: a simple, versatile, and robust algorithm for endocrine pulse detection. Am J Physiol 1986;250:E486-93.
1530 Fujimoto at al.
12. Sollenberger MJ, Carlsen EC, Booth RAJr, Johnson ML, Veldhuis JD, Evans WS. Nature of gonadotropin-releasing hormone self-priming of luteinizing hormone secretion during the normal menstrual cycle. AM J OBSTET GYNECOL 1990; 163:1529-34. 13. Quigley M, Yen S. The role of endogenous opiates on LH secretion during the menstrual cycle. J Clin Endocrinol Metab 1980;51:179-81. 14. Ropert J, Quigley M, Yen S. Endogenous opiates modulate pulsatile luteinizing hormone release in humans. J Clin Endocrinol Metab 1981;52:583-5. 15. Nicoletti I, Filipponi P, Sfrappini M. Catecholamines and pituitary function. I. Effects of catecholamine synthesis inhibition and subsequent catecholamine infusion on gonadotropin and prolactin serum levels in normal cycling women and in women with hyperprolactinemic amenorrhea. Horm Res 1984;19:158-70. 16. Nicoletti I, Filipponi P, Fedeli L, Angeletti G, Santeusanio F, Brunetti P. Effect of methyl-p-tyrosine administration on gonadotropin and prolactin release in normal-cycling, postmenopausal, and hyperprolactinemic women. Acta Eur Fertil 1981;12:271-3. 17. Cetel N, Quigley M, Yen S. Naloxone-induced prolactin secretion in women: evidence against a direct prolactin stimulatory effect of endogenous opioids. J Clin Endocrinol Metab 1985;60:191-6. 18. Judd S, Rakoff J, Yen S. The effects of ovariectomy and estrogen treatment on the dopamine inhibition of gonadotropin and prolactin release. J Clin Endocrinol Metab 1979;49: 182-4. 19. Bhattacharya A, Dierschke D, Yamaji J, Knobil E. The pharmacologic blockade of the circhoral mode of LH secretion in the ovariectomized rhesus monkey. Endocrinology 1972;90:778-86. 20. Crowley W, Terry L, Johnson M. Evidence for the involvement of central epinephrine systems in the regulation of luteinizing hormone, prolactin, and growth hormone release in female rats. Endocrinology 1982; 110: 1102-7. 21. Heijna M, Padt M, Hogenboom F, Schoffelmeer A, Mulder
December 1993 Am J Obstet Gynecol
A. Opioid-receptor-mediated inhibition of [3H]dopamine
22.
23.
24.
25.
26. 27. 28.
29. 30.
but not [3H]noradrenaline release from rat mediobasal hypothalamus slices. Neuroendocrinology 1991 ;54: 11826. Leadem C, Crowley W, Simpkins J, Kalra S. Effects of naloxone on catecholamine and LHRH release from the perifused hypothalamus of the steroid-primed rat. Neuroendocrinology 1985;40:497-500. Rasmussen D, Liu J, Wolf P, Yen S. Neurosecretion of human hypothalamic immunoreactive l3-endorphin: in vitro regulation by dopamine. Neuroendocrinology 1987; 45:197-200. Sarkar D, Yen S. Hyperprolactinemia decreases the luteinizing hormone-releasing hormone concentration in pituitary portal plasma: a possible role for l3-endorphin as a mediator. Endocrinology 1985;116:2080-4. Goldsmith P, Thind K. Modulation of GnRH pulse generator activity. In: Genazzani AR, Montemagno D, Nappi C, Petraglia F, eds. The brain and female reproductive function. Lancaster, England: Parthenon, 1987:21-33. Wehrenberg W, Wardlaw S, Frantz A, Ferin M. I3-Endorphin in hypophyseal portal blood: variations throughout the menstrual cycle. Endocrinology 1982; 111:879-81. Wardlaw S, Wehrenburg W, Ferin M, Antunes J, Frantz A. Effect of sex steroids on l3-endorphin in hypophyseal portal blood. J Clin Endocrinol Metab 1982;55:877-81. Kohama S, Freesh F, Bethea C. Immunocytochemical colocalization of hypothalamic progestin receptors and tyrosine hydroxylase in steroid-treated monkeys. Endocrinology 1992;131:509-17. Barnes R, Lobo R. Central opioid activity in polycystic ovary syndrome with and without dopaminergic modulation. J Clin Endocrinol Metab 1985;61:779-82. Melis G, Cagnacci A, Gambacciani M, Paoletti A, Moggi L, Fioretti P. Restoration ofluteinizing hormone response to naloxone in postmenopausal women by chronic administration of the anti-dopaminergic drug veralipride. J Clin Endocrinol Metab 1988;66:964-7.