Estradiol requirements for induction and maintenance of the gonadotropin-releasing hormone surge: implications for neuroendocrine processing of the estradiol signal.

Two experiments were performed to examine the temporal requirements of the estradiol signal for the GnRH and LH surges in the ewe. Hypophyseal portal and jugular blood (to measure GnRH and LH, respectively) were sampled from ewes set up in an artificial follicular phase model. After progesterone withdrawal to simulate luteolysis, circulating estradiol was raised to a preovulatory level by inserting estradiol implants, which then were removed at different times to vary estradiol signal duration. The objective of the first experiment was to assess the effect of withdrawing estradiol at surge onset on development and maintenance of the GnRH/LH surges. Removal of estradiol, before surge onset, neither altered the LH surge in relation to that induced when the estradiol stimulus was maintained nor affected stimulation of a massive and sustained GnRH surge that outlasted the LH surge by many hours. Continued estradiol treatment, however, did prolong the GnRH surge. In the second experiment, the estradiol stimulus was shortened to test the hypothesis that estradiol need not be present for the whole presurge period to induce GnRH/LH surges. Ewes received estradiol either up to the time of surge onset (21 h) or for periods equivalent to the last 14 h, the last 7 h, or the earliest 7 h of the 21-h signal. Shortening the signal to 14 h did not reduce its ability to stimulate a full GnRH surge, but it did reduce the amplitude of the resultant LH surge. Further shortening of the signal to 7 h, however, produced a mixed response. Most animals (8 of 10 combining the two 7-h groups) did not express GnRH surges. In the two ewes that did, GnRH surge amplitude and duration were again within the range observed with the 21-h estradiol signal, but the LH response was greatly reduced. These results indicate that, once the GnRH/LH surges of the ewe have begun, elevated estradiol is not required for surge maintenance. Development of a full GnRH surge requires elevated estradiol for only a portion of the presurge period. More prolonged exposure to estradiol, however, is needed to maximize pituitary responsiveness to GnRH. Since the estradiol signal for the GnRH surge is relatively short (7-14 h) and temporally located well in advance of the surge itself, these results are consistent with the hypothesis that estradiol is required only to activate the steroid-responsive neuronal elements and not for progression of the signal from these elements to the actual surge process of GnRH release.