It is well known that estrogen (E2) stimulates expression of progesterone receptors (PRs), thereby inducing responsiveness of several tissues to the actions of progesterone (P). Recent studies have also suggested, however, that biological actions previously ascribed to E2 alone may also be mediated by activation of E2-induced PRs, even independently of signal changes in P concentrations. In the present experiments, the progesterone receptor knockout (PRKO) mice were used to assess the role of PR activation in the positive feedback actions of E2 on gonadotropin release. Ovariectomized (OVX) PRKO mice were tested for their capacity to mount primary gonadotropin surges in response to exogenous E2, and to exhibit a GnRH self-priming effect in response to sequential injections of the decapeptide. Wild-type (WT) and PRKO mice were OVX, treated with both 17beta-estradiol and estradiol benzoate (EB), and then killed at 1900 h on day 7 postOVX. Plasma LH RIA revealed that WT mice exhibited surges in response to the E2 treatment; the PRKO mice, however, showed no elevation in plasma LH above untreated controls. Instead, plasma LH levels in E2-treated, OVX PRKO mice decreased significantly in comparison to untreated OVX PRKO mice, suggesting that E2 can exert a negative feedback influence on LH release in PRKO mice, despite the absence of positive feedback effects. A slight but significant rise in plasma FSH was observed in E2-treated OVX WT mice in comparison to untreated controls: an effect not seen in E2-treated OVX PRKO mice, reinforcing the observation that estrogen's positive feedback effects are compromised in PRKO mice. In a second experiment, E2-treated OVX WT and PRKO mice were given either one or two pulses of GnRH 60 min apart, and killed 10 min later. The WT mice were found to exhibit a robust GnRH self-priming effect, as WT mice receiving two GnRH pulses displayed LH responses approximately 2-fold greater than those receiving only one pulse. By contrast, PRKO mice receiving two GnRH pulses exhibited no additional increase in plasma LH levels. We conclude that PR activation is obligatory for expression of the GnRH self-priming effect as well as for generation of E2-induced LH and FSH surges. The extent to which failure of LH surge secretion in PRKO mice is due to the absence of GnRH self-priming, lack of hypothalamic GnRH surges, and/or defects in other processes remains to be determined. These observations clearly demonstrate, however, that the presence of PR is an absolute requirement for the transmission of E2-induced signals leading to gonadotropin surges.