| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Division of Endocrinology, Department of Internal Medicine (C.R.M., W.S.E., J.C.M.); Center for Research in Reproduction (C.R.M., M.B.G., Y.H., W.S.E., J.C.M.); and Department of Obstetrics and Gynecology (W.S.E.), University of Virginia Health System, Charlottesville, Virginia 22908
Address all correspondence and requests for reprints to: Christopher R. McCartney, M.D., Box 800746, Division of Endocrinology, Department of Internal Medicine, University of Virginia Health System, Charlottesville, Virginia 22908. E-mail: . cm2hq{at}virginia.edu
Abstract
The luteal-follicular transition is characterized by decreasing plasma levels of E2, progesterone (P), and inhibin A, with concomitant increases in FSH and LH levels. LH (and by inference GnRH) pulse frequency increases from 1 pulse every 3–4 h during the luteal phase to approximately 1 pulse/h at the midcycle LH surge. To examine the regulation of GnRH pulse frequency, we gave 10 normally cycling women transdermal E2 and oral P to produce midluteal levels [364 ± 65.0 pmol/liter (99 ± 18 pg/ml) and 29.7 ± 6.8 nmol/liter (9.3 ± 2.1 ng/ml), respectively] for 10 d after the LH surge (d 0). P was then discontinued, and E2 was given alone for 3 additional wk. Pulsatile LH secretion and follicular size were assessed on d 10, 17, 24, and 31. Results are presented as the mean ± SEM.
LH pulse frequency was 3.1 ± 0.5 pulses/12 h after 10 d of E2 and P, and remained low on d 17 when P had fallen below 1.6 nmol/liter (<0.5 ng/ml). In the continued presence of midluteal levels of E2 [
360 pmol/liter (100 pg/ml)], LH pulse frequency increased on d 24 and 31 to 5.5 ± 0.9 and 5.8 ± 0.5 pulses/12 h, respectively, whereas pulse amplitude remained unchanged. FSH increased 2-fold, but follicular size did not change.
These results are consistent with E2 potentiating the effects of low concentrations of P on the GnRH pulse generator for at least 7 d, after which pulse frequency increases despite maintenance of E2 levels. This supports the hypothesis that the increasing GnRH pulse frequency throughout the follicular phase reflects the gradual loss of the inhibitory actions of low concentrations of P.
GnRH IS SECRETED in a pulsatile fashion by a specialized network of hypothalamic neurons termed the GnRH pulse generator. Pulsatile secretion of GnRH stimulates both LH and FSH synthesis and secretion by pituitary gonadotropes. The relative amounts of secreted LH and FSH vary throughout the human menstrual cycle. Preferential secretion of FSH occurs during the early to midfollicular phase; this is important for follicular development and induction of granulosa cell LH receptors and aromatase activity. Throughout the remainder of the follicular phase, LH secretion predominates. Differential regulation of LH and FSH secretion is achieved via at least two mechanisms. Gonadal steroids and inhibins exert specific feedback actions on the gonadotrope to augment LH and/or inhibit FSH secretion. In addition, a changing pattern of GnRH pulse secretion appears to play an important role. In both rodent (1) and primate (2) models, faster GnRH pulse frequencies favor LH, whereas slower frequencies favor FSH synthesis and secretion. Thus, the slowing of GnRH pulse frequency during the luteal phase is one mechanism involved in enhancing FSH synthesis and subsequent secretion, and as such may be important for the long-term maintenance of cyclic ovulation.
The mechanisms controlling LH (and by inference GnRH) pulse frequency throughout the human menstrual cycle are unclear. Progesterone (P) appears to be the predominant hormone mediating the luteal slowing of GnRH pulse frequency, as evidenced by the ability of P to slow LH pulse frequency when administered to women during the follicular phase (3). P similarly decreases GnRH pulse frequency in sheep in the presence of E2 (4). However, the role of E2 alone in the regulation of GnRH pulse frequency remains unclear. E2 decreases LH pulse frequency temporally in postmenopausal women (5) and also reduces LH pulsatility and hypothalamic electrical activity in ovariectomized monkeys (6), suggesting that E2 suppresses GnRH neuronal activity. In contrast, E2 alone has no effect on LH pulse frequency in sheep (4), and LH pulse frequency is high in the late follicular phase in women (7, 8), a time when E2 levels are elevated.
Nippoldt et al. (9) investigated the relative roles of E2 and P in the luteal slowing of GnRH pulse frequency in normal women. E2 alone, P alone, or E2 plus P was begun at the midluteal phase and continued for 8.7 d, on the average. Administration of E2 alone, while allowing P levels to fall, maintained a slow LH pulse frequency, whereas maintenance of P alone did not prevent the expected increase in LH pulse frequency during the luteal-follicular transition. These data suggest that either E2 directly inhibits GnRH pulse frequency, or that E2 can potentiate or maintain the effects of low concentrations of P. The latter may involve E2 maintaining hypothalamic P receptors (10, 11) allowing continuing sensitivity to low concentrations of P.
In the current study we aimed to elucidate the relative roles of luteal steroids in controlling GnRH pulse frequency during the transition from the luteal to the late follicular phase. We postulated that the luteal slowing of GnRH pulse frequency is effected by P acting to increase hypothalamic opioid activity (12, 13, 14, 15) and that the gradual increase in GnRH pulse frequency from the early to the late follicular phase reflects the gradual loss of the restraining actions of P. Specifically, we tested the hypothesis that if midluteal E2 concentrations were maintained for 3 wk after the luteal phase while allowing P levels to fall, GnRH pulse frequency would eventually increase, suggesting that E2 alone is insufficient to maintain a slow LH pulse frequency.
Subjects and Methods
Subjects
Ten healthy women (aged 19–35 yr) were studied. All women were of normal weight for height (body mass index, 18.3–26.0 kg/m2), had regular menstrual cycles (every 24–32 d), and had no evidence of hyperandrogenism. All subjects were screened with determinations of LH, FSH, E2, estrone, P, total T, 17-hydroxyprogesterone, dehydroepiandrosterone sulfate, fasting insulin and glucose, PRL, T4, and hCGß. The women were not taking any medications known to affect the reproductive axis and had not taken any hormonal medications for at least 90 d before study screening.
Study protocol
The study was approved by both the human investigation committee of the University of Virginia Health System and the General Clinical Research Center advisory committee. Informed consent was obtained from all study volunteers.
Evidence for ovulation was obtained in a control cycle during which women used urinary LH detection kits (Clear and Easy, Unipath Diagnostics Co., Princeton, NJ) to approximate the timing of the LH surge. On the day of initial urinary LH test positivity, designated control cycle d 0, a blood sample for plasma LH, FSH, E2, and P determinations was obtained. These (single) LH determinations did not necessarily reflect peak LH values during a surge, but were obtained as evidence that an LH surge was in progress. Subsequent measurements of E2 and P were obtained on d 3 and 7 to assess whether ovulation had occurred.
During the subsequent menstrual cycle (study cycle), women again used urinary LH detection kits to approximate the LH surge (confirmed with plasma LH, FSH, E2, and P), and this day was similarly designated study cycle d 0. E2 patches (two patches, each patch delivering 0.1 mg/d for a total of 0.2 mg/d, changed every 3 d; Estraderm, Novartis Pharmaceuticals, East Hanover, NJ) and oral micronized P (100 mg every 8 h; Prometrium, Solvay Pharmaceuticals, Inc., Marietta, GA) were started within 24 h of the LH surge and were continued through the first General Clinical Research Center admission.
On study cycle d 8–12 (designated study d 10), LH pulse frequency and ovarian follicular size were assessed with frequent blood sampling and transvaginal ultrasound, respectively. Subjects were admitted no later than 1800 h, 2 h before sampling, and blood samples were obtained through an indwelling iv forearm heparin lock over a 13-h period as follows: samples for LH every 10 min, for FSH every 60 min, and for E2 and P every 2 h. GnRH (25 ng/kg) was given iv 1 h before completion of frequent sampling to evaluate pituitary LH responses to GnRH. For nine subjects, aliquots of samples from 2200–2240 h and from 0600–0640 h were pooled for measurements of inhibin A and inhibin B (i.e. two pooled samples per admission for each peptide); insufficient sample quantity prohibited measurement of inhibins in one subject. After the first General Clinical Research Center admission, patients discontinued oral P, but continued transdermal E2 patches. Subjects were studied weekly thereafter for 3 wk (i.e. d 17, 24, and 31), as described above (see Fig. 1
). Oral iron supplementation (325 mg, twice daily) was given throughout the study.
|
All samples from an individual woman were analyzed in duplicate in the same assay for each hormone. Plasma LH and FSH were measured by chemiluminescence [Nichols Institute Diagnostics, San Juan Capistrano, CA; assay sensitivities, 0.01 and 0.02 IU/liter, respectively; intraassay coefficients of variation (CV), 11% and 6%; interassay CV, 10% and 10%, respectively]. E2 and P were measured by RIA [Diagnostics Systems Laboratories, Inc., Webster, TX; sensitivities were 17.2 pmol/liter (4.7 pg/ml) and 0.38 nmol/liter (0.12 ng/ml); intraassay CV, 8% and 9%; interassay CV, 14% and 18%, respectively]. T, estrone, and insulin were also measured by RIA [Diagnostics Systems Laboratories, Inc.; sensitivities were 0.28 nmol/liter (0.08 ng/ml), 4.4 pmol/liter (1.2 pg/ml), and 9.3 pmol/liter (1.3 µIU/ml); intraassay CV, 7%, 8%, and 6%; interassay CV, 15%, 14%, and 12%, respectively]. Inhibins A and B were measured by ELISA (Serotec, Oxford, UK). The sensitivities of the inhibin A and inhibin B assays were 0.6 IU/ml and 15.6 pg/ml, respectively (both representing values of the lowest calibrator). For inhibin A quality controls of 2.8 and 22 IU/ml, the intraassay CV were 3% and 7%, and the interassay CV were 7.0% and 7.2%, respectively. The inhibin A assay is standardized as equivalents of the First International Standard for inhibin, WHO recombinant human inhibin (identification no. 91/624), received from National Institute for Biological Standards and Controls (Hertfordshire, UK). For inhibin B, the intraassay CV ranged from 4–6%, and the interassay CV ranged from 15–18% for serum spiked with 121, 250, and 723 pg/ml inhibin B. The inhibin B measurements are based upon nominal mass of the calibrators provided by the manufacturer. Samples with measured values below assay sensitivity were assigned the value of the assay’s sensitivity.
Data analysis
The 10-min samples were assayed for LH, with the resulting luminometer output data constructed using a variable weighting hormone concentration data reduction protocol developed by Straume et al. (16). This procedure assigns statistically accurate estimates of unknown hormone concentrations, with associated uncertainties, based on experimental uncertainties in sample replicates and the fitted calibration curve. Experimental error is addressed by assigning and propagating uncertainty estimates for each measured standard curve response (including zero dose responses) by an empirically determined discrete uncertainty profile and by propagating calibration curve uncertainty. Discrete uncertainty profiles account for both response precision (replicability) and accuracy (deviation from the predicted calibration curve) without relying on assumed theoretical variance-assay response relationships.
LH pulses were identified using the computer algorithm Cluster (Cluster 7). The parameters used for analysis were a test nadir and peak size of 2 x 2 with a t statistic of 2 for both the upstroke and the downstroke (17). Amplitudes were calculated as the peak LH minus the preceding nadir as determined by the Cluster 7 program. Missing values represented less than 0.1% of the total and were ignored.
If the amplitude of an LH pulse detected by the Cluster 7 program was less than the range of intraassay variability for the LH chemiluminescence method, it was not considered a pulse in subsequent analysis. The ranges of intraassay LH variability were established by repeated measurements of LH (i.e. 20 measurements) on 3 individual samples with LH concentrations spanning those seen in normal women (i.e. 0.48, 2.41, and 9.98 IU/liter). Repeated measurements yielded values with a range (highest value minus lowest value) of 0.24, 0.48, and 0.93 IU/liter, respectively. Therefore, the following pulses detected by Cluster were excluded from further analysis: peak less than 1 IU/liter with an amplitude below 0.25 IU/liter; peak of 1 or more and less than 5 IU/liter, amplitude less than 0.5 IU/liter; and peak of 5 IU/liter or more and amplitude less than 1.0 IU/liter.
Statistical methods
Results are presented as the mean ± SEM. With the exception of LH pulse frequency, all data were transformed to the natural logarithmic scale before analysis to stabilize residual variation across time periods. Inhibin values that fell below the minimum detectable level were set equal to the detectable limit.
The between time period comparisons of those outcomes that were analyzed on the logarithmic scale are presented in terms of the ratio of the geometric means (18). The geometric mean is a location parameter, similar to the arithmetic mean and median, and is calculated by simply taking the antilogarithm of the mean response computed from the logarithmically transformed data. For the LH pulse frequency we have presented the between time period comparisons in terms of the change in the arithmetic mean of the response.
All data other than the inhibin A data were analyzed by mixed effect ANOVA (19). The inhibin A data were analyzed by paired t test.
Model specification for the mixed effect ANOVA was designed to compare the mean within-subject response on d 17, 24, and 31 to the mean response on d 10. ANOVA model parameters were estimated by residual maximum likelihood, and the within-subject variance-covariance matrix was modeled in the form of the compound symmetry variance-covariance matrix. The multiple comparison adjustment was based on Dunnett’s criterion with an experimentwise type I error rate of 0.05.
All statistical computations were carried out in SAS version 8.0 with software of PROC MIXED (SAS Institute, Inc., Cary, NC) (19).
Results
All subjects had normal screening laboratory results, and volunteer weights remained stable throughout the study (data not shown). On d 0 of the control cycle, corresponding to 16.7 ± 1.1 d after the beginning of the previous menstrual period, hormonal measurements were as follows: LH, 38.2 ± 7.8 IU/liter; FSH, 12.6 ± 2.7 IU/liter; E2, 470 ± 75 pmol/liter (128 ± 20.4 pg/ml); and P, 5.7 ± 1.6 nmol/liter (1.8 ± 0.5 ng/ml). On control d 3 and 7, E2 concentrations were 255 ± 37.4 pmol/liter (69.6 ± 10.2 pg/ml) and 383 ± 63.9 pmol/liter (104.4 ± 17.4 pg/ml); P concentrations were 21.3 ± 5.1 nmol/liter (6.7 ± 1.6 ng/ml) and 38.8 ± 10.5 nmol/liter (12.2 ± 3.3 ng/ml), respectively. On d 0 of the study cycle, corresponding to 15.7 ± 1.5 d after the beginning of the previous menstrual period, hormonal measurements were as follows: LH, 57.0 ± 11.0 IU/liter; FSH, 15.1 ± 3.8 IU/liter; E2, 748 ± 77.4 pmol/liter (204 ± 21.1 pg/ml); and P, 3.8 ± 0.3 nmol/liter (1.2 ± 0.1 ng/ml).
Steroids
Gonadal steroid levels are shown in Fig. 2. Plasma E2 on d 10 of the study cycle was 363 ± 66.1 pmol/liter (99 ± 18 pg/ml) and was not statistically different during subsequent study days; likewise, values were similar to plasma E2 concentrations measured during the luteal phase of the control cycle. The plasma P concentration on d 10 was 29.6 ± 6.7 nmol/liter (9.3 ± 2.1 ng/ml) and was similar to the level during the control luteal phase. After withdrawal of oral P, plasma P fell to 1.3 ± 0.1 nmol/liter (0.4 ± 0.03 ng/ml) by d 17 and remained low [<1.3 nmol/liter (<0.4 ng/ml)] throughout the remainder of the study (P < 0.001 for d 10 vs. all other admissions).
|
FSH, inhibin A, and inhibin B data are shown in Fig. 3. Plasma FSH increased progressively through the study and had increased 2.1-fold by d 31 (P < 0.01).
|
Inhibin A levels also varied, being measurable in five of nine subjects on d 10. In contrast to inhibin B, inhibin A levels were below assay sensitivity [<0.6 IU/ml (<4 pg/ml)] in all subjects on d 17, 24, and 31.
LH
After 10 d of E2 and P administration, mean LH pulse frequency was 3.1 ± 0.53 pulses/12 h, and remained low (3.4 ± 0.56 pulses/12 h) on d 17 when P had fallen below 1.6 nmol/liter (<0.5 ng/ml; Fig. 4). Subsequently, in the presence of constant levels of E2, LH pulse frequency increased to 5.5 ± 0.92 pulses/12 h by d 24 (P < 0.005 compared with d 10) and 5.8 ± 0.47 pulses/12 by d 31 (P = 0.001 compared with d 10). The mean LH pulse amplitude was 1.86 ± 0.42 IU/liter on d 10 and did not change significantly throughout the study. The mean LH level rose on d 24, but only achieved statistical significance on d 31 (2.2-fold increase; P = 0.01 compared with d 10). The LH response to GnRH was 7.9 ± 1.4 IU/liter on d 10; this response did not change significantly throughout the remainder of the study.
|
Ovarian follicular size did not change significantly throughout the study (P > 0.05). Mean follicular size was 18.3 ± 6.9 mm on d 10. In 6 of 10 subjects, no dominant follicle was seen by ultrasound. However, a dominant cystic structure was seen in 4 individuals on d 10; these presumably represented unruptured follicles. These dominant cysts uniformly decreased in size over time and disappeared under surveillance in all but 1 subject. Mean cyst size progressively fell to 10.4 ± 3.7 mm by d 31.
Discussion
The patterns of LH, and by inference GnRH, pulsatile secretion vary during an ovulatory menstrual cycle, and consistent findings in women have been reported by several groups (7, 8, 20, 21). However, the mechanisms regulating GnRH pulse secretion are incompletely understood, and the relative contributions of the ovarian steroids E2 and P are not clearly defined. P is generally accepted to be the main arbiter of the decreased GnRH pulse frequency during the luteal phase, based on earlier studies in women and sheep (3, 4). In these studies P was administered when E2 concentrations were also elevated, and it remained uncertain whether E2, P, or both steroids are necessary for slowing of GnRH pulse frequency. Indeed, E2 slows GnRH pulse frequency temporally in both postmenopausal women (5) and ovariectomized rhesus monkeys (6), and antiestrogens increase LH pulse frequency when given to women during the luteal phase (22). However, the mid to late follicular increase of plasma E2 does not suppress GnRH frequency in the presence of low concentrations of P, and the late follicular phase is characterized by the fastest LH pulse frequency seen in cycling women. In addition, E2 is required to achieve maximal GnRH pulse frequencies during the follicular phase in sheep (23).
The current investigation was pursuant to an earlier study (9) that demonstrated that maintaining P levels into the follicular phase, while allowing E2 levels to fall, did not prevent the expected increase in LH pulse frequency. However, when plasma E2 was similarly maintained into the early follicular phase, while permitting P levels to fall, a slow LH pulse frequency of 4.3 ± 0.6 pulse/10 h was maintained. These unexpected results suggested that either E2 directly inhibits GnRH pulse frequency, or that E2 can potentiate or maintain the effects of low concentrations of P. Synergistic effects of E2 and P on GnRH pulse frequency have been suggested in other studies. For instance, in recently ovariectomized sheep, a P concentration of 15 ± 1.6 nmol/liter (4.7 ± 0.5 ng/ml) decreased LH pulse frequency, whereas 4 ± 0.6 nmol/liter (1.3 ± 0.2 ng/ml) was only effective in the presence of E2 (4). The apparent effect of E2 to potentiate the biological actions of P may be mediated by the ability of E2 to induce and/or maintain an increased number of hypothalamic PR (10, 11). We assessed whether the apparent effect of E2 to maintain a slow GnRH pulse frequency (9) reflected amplification of the actions of low levels of P. To this end, we continued E2 alone for an additional 3 wk after combined E2 and P administration, assessing changes in LH pulse frequency.
In an effort to standardize hypothalamic exposure to gonadal steroids, E2 and P were given to all subjects within 24 h of the LH surge, and midluteal plasma concentrations were achieved. As anticipated, the presence of P was associated with slowing of LH pulses to luteal frequencies, and E2 levels of 290–370 pmol/liter (80–100 pg/ml) maintained a slow LH pulse frequency for 7 d after withdrawal of P, confirming earlier findings using less physiological E2 concentrations [i.e. approximately 990-1030 pmol/liter (270–280 pg/ml)] (9). Despite continued maintenance of midluteal E2 concentrations, the LH pulse frequency subsequently increased approximately 1.8-fold by 21 d after withdrawal of P. LH pulse amplitude did not change throughout the study, suggesting that the increase in LH pulse frequency alone accounts for the increase in mean LH seen 21 d after P withdrawal.
The steroid profiles achieved during the first 10 d of our protocol do not reproduce those seen in normal women during the luteal phase. We therefore did not include a control group of cycling women who did not take E2 and P, as they would not have been exposed to the same steroid milieu. In addition, the aim was not specifically to make a direct comparison to normal women during the luteal-follicular transition, but to assess the effect of continued E2 administration while allowing P to fall. Regardless, one may compare our results to historical controls. In normal women, LH pulse frequency increases in a relatively rapid manner over the luteal-follicular transition, increasing 4.5-fold during an 8-d period (24). Conversely, in our experimental group the increase in LH pulse frequency was blunted and delayed by approximately 1 wk.
We and others have suggested that after pubertal maturation the human GnRH pulse generator has an inherent maximal firing frequency of approximately one pulse per h in the absence of ovarian restraint (25, 26, 27). In this model any reduction in this inherent (or basal) pulse frequency occurs as a result of inhibition by gonadal steroids, and the initial increase in GnRH pulse frequency observed during the luteal-follicular transition reflects the abrupt decreases in E2 and P concentrations. In a similar vein, the gradual increase in GnRH pulse frequency observed from the early to the late follicular phase may be a result of the gradual loss of the restraining effects of low levels of P. The continued actions of low levels of P during the first part of the follicular phase may be facilitated via the maintenance of hypothalamic P receptors induced by increasing E2 secretion during the follicular phase. The final loss of the E2-potentiated effects of low levels of P would subsequently occur within approximately 2 wk. The current data support such a hypothesis.
However, the degree and time course of changes in GnRH pulse secretion in the present study and during ovulatory cycles are not in full accord. LH pulse frequency on d 31 (5.8 pulses/12 h) was somewhat slower than those observed in most studies at the midcycle LH surge. The reason(s) for this difference is unclear, but may reflect plasma E2 profiles that by design differed from those seen in normal cycles. In contrast to the rapid fall in E2 concentrations during the late luteal phase, maintaining midluteal E2 concentrations may have delayed or prolonged the recovery of GnRH pulse frequency. The continued administration of E2 could have maintained a sufficient number of hypothalamic PR to allow ongoing responses to very low levels of P, retarding effective withdrawal of hypothalamic exposure to P, and therefore the increase in GnRH pulse frequency. Thus, the length of surveillance in our study may not have been long enough to observe an increase to maximal pulse frequencies. An inability to detect pulses on the basis of decreased pituitary responses to GnRH is unlikely, as LH responses to a low physiological dose of GnRH (25 ng/kg) were easily detectable in all subjects during all admissions.
An alternative explanation for these findings is that frequent blood sampling was performed overnight, as sleep-entrained slowing of LH pulse frequency during the first half of the follicular phase of the human cycle has been described (8, 28). The mechanisms of this phenomenon are unknown and may in part explain the slower GnRH pulse frequency on d 17 of the present study, but are unlikely to account for later estimates of pulse frequency, as the nocturnal slowing is not seen during the normal late follicular phase. It should be noted, however, that this would not explain the previously described results of Nippoldt et al. (9), as LH pulse frequency was assessed during the day in this study.
It remains possible that LH pulse frequency would increase regardless of continued administration of E2 and P. However, prior work indicates that continued E2 and P would maintain a slow LH pulse frequency. Previous studies in both normal women and women with polycystic ovary syndrome, a disorder marked by persistently rapid LH pulses, document that administration of E2 and P or combined oral contraceptive pills for approximately 3 wk maintains a slow LH pulse frequency (29, 30).
Mean FSH increased with each admission, achieving statistical significance by d 24. This increase in FSH probably reflects the actions of increasing GnRH pulse frequency in the presence of increased pituitary FSH stores, as the prior slow GnRH pulse stimulus favors FSH synthesis (1, 2). The importance of increasing GnRH pulse frequency to the rise of FSH was also highlighted in a study using 1 of 2 regimens of exogenous pulsatile GnRH in GnRH-deficient women (31). In this study increasing GnRH pulse frequency from every 4 h to every 90 min on the day of menses resulted in a rise in FSH that did not differ from that in normally cycling women, whereas continuing a slow GnRH pulse frequency (i.e. one pulse every 4 h) for 6 d into the follicular phase resulted in lower FSH levels. In the present study the increasing GnRH pulse frequency appeared to overcome direct inhibition of FSH secretion by E2 levels of 290–370 pmol/liter (80–100 pg/ml); however, FSH secretion appeared to have been partially restrained compared with that during a normal luteal-follicular transition (32). It is difficult to interpret the inhibin A and inhibin B results due to their variability among individual subjects, with many values below assay sensitivity. As the main source of inhibin B appears to be from developing follicles (33), and follicular development was not detected by transvaginal ultrasound, the frequent occurrence of undetectable inhibin B levels might be expected. Inhibin A levels were detectable in 6 of 10 subjects on study d 10 and were subsequently below assay sensitivity in all subjects, which parallels inhibin A profiles during normal cycles. Inhibin A originates from the corpus luteum, and our inability to detect inhibin A levels in 4 subjects on study d 10 suggests impaired luteal function resulting from partial gonadotropin suppression by exogenous luteal steroids. Thus, the rise in plasma FSH reflects the actions of increasing GnRH secretion, with FSH release partially restrained by the continuing presence of E2, so that follicular development was not seen during the study period.
In summary, we conclude that E2 potentiates the effect of low concentrations of P on the GnRH pulse generator for more than 7 d, after which GnRH pulse frequency increases despite maintenance of midluteal concentrations of E2. These data support the hypothesis that the increase in GnRH pulse frequency observed throughout the follicular phase of the menstrual cycle reflects the gradual loss of the inhibitory actions of low concentrations of P and support a dominant role for P in regulating the hypothalamic pulse generator during ovulatory cycles.
Acknowledgments
We gratefully acknowledge the following: the nurses and support staff of the General Clinical Research Center at the University of Virginia for their expert implementation of the protocol described herein; Michael L. Johnson, Ph.D., for providing the pulse detection program Cluster 7 and for his expert technical assistance; Patrick M. Sluss, Ph.D. (Reproductive Endocrine Unit, Massachusetts General Hospital), for his timely and expert assays of inhibins A and B; James T. Patrie (Department of Health Evaluation Sciences, Division of Biostatistics and Epidemiology, University of Virginia Health System) for statistical evaluation of study data; and Martin Straume, Ph.D., for assistance with the variable weighting hormone concentration data reduction protocol used herein.
Footnotes
This work was supported by the NICHHD, NIH, through Cooperative Agreement U54-HD-28934 as part of the Specialized Cooperative Centers Program in Reproduction Research, Grant HD-34179 (to J.C.M.), by General Clinical Research Center Grant M01-RR-00847, and by NIH Training Grant T32-HD-07382 (to C.R.M.).
Abbreviations: CV, Coefficient of variation; P, progesterone.
Received September 19, 2001.
Accepted February 5, 2002.
References
This article has been cited by other articles:
 |
|
 |
|
  Y. Tenenbaum-Rakover, M. Commenges-Ducos, A. Iovane, C. Aumas, O. Admoni, and N. de Roux Neuroendocrine Phenotype Analysis in Five Patients with Isolated Hypogonadotropic Hypogonadism due to a L102P Inactivating Mutation of GPR54 J. Clin. Endocrinol. Metab., March 1, 2007; 92(3): 1137 - 1144. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. R. McCartney, S. K. Blank, and J. C. Marshall Progesterone acutely increases LH pulse amplitude but does not acutely influence nocturnal LH pulse frequency slowing during the late follicular phase in women Am J Physiol Endocrinol Metab, March 1, 2007; 292(3): E900 - E906. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Janovick, S. P. Brothers, P. E. Knollman, and P. M. Conn Specializations of a G-protein-coupled receptor that appear to aid with detection of frequency-modulated signals from its ligand FASEB J, February 1, 2007; 21(2): 384 - 392. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. E. Messinis Ovarian feedback, mechanism of action and possible clinical implications Hum. Reprod. Update, September 1, 2006; 12(5): 557 - 571. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S.K. Blank, C.R. McCartney, and J.C. Marshall The origins and sequelae of abnormal neuroendocrine function in polycystic ovary syndrome Hum. Reprod. Update, July 1, 2006; 12(4): 351 - 361. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Pielecka, S. D. Quaynor, and S. M. Moenter Androgens Increase Gonadotropin-Releasing Hormone Neuron Firing Activity in Females and Interfere with Progesterone Negative Feedback Endocrinology, March 1, 2006; 147(3): 1474 - 1479. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chhabra, C. R. McCartney, R. Y. Yoo, C. A. Eagleson, R. J. Chang, and J. C. Marshall Progesterone Inhibition of the Hypothalamic Gonadotropin-Releasing Hormone Pulse Generator: Evidence for Varied Effects in Hyperandrogenemic Adolescent Girls J. Clin. Endocrinol. Metab., May 1, 2005; 90(5): 2810 - 2815. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. Sullivan and S. M. Moenter GABAergic Integration of Progesterone and Androgen Feedback to Gonadotropin-Releasing Hormone Neurons Biol Reprod, January 1, 2005; 72(1): 33 - 41. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Dafopoulos, C.G. Kotsovassilis, S. Milingos, A. Kallitsaris, G. Galazios, E. Zintzaras, P. Sotiros, and I.E. Messinis Changes in pituitary sensitivity to GnRH in estrogen-treated post-menopausal women: evidence that gonadotrophin surge attenuating factor plays a physiological role Hum. Reprod., September 1, 2004; 19(9): 1985 - 1992. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Eagleson, A. B. Bellows, K. Hu, M. B. Gingrich, and J. C. Marshall Obese Patients with Polycystic Ovary Syndrome: Evidence that Metformin Does Not Restore Sensitivity of the Gonadotropin-Releasing Hormone Pulse Generator to Inhibition by Ovarian Steroids J. Clin. Endocrinol. Metab., November 1, 2003; 88(11): 5158 - 5162. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. K. Welt, Y. L. Pagan, P. C. Smith, K. B. Rado, and J. E. Hall Control of Follicle-Stimulating Hormone by Estradiol and the Inhibins: Critical Role of Estradiol at the Hypothalamus during the Luteal-Follicular Transition J. Clin. Endocrinol. Metab., April 1, 2003; 88(4): 1766 - 1771. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
