We have recently shown that during the diurnal peak mineralocorticoid receptors (MR) can rapidly (<25 min) regulate basal corticosterone (CORT) concentrations (Atkinson et al. 2006). In the current study we examined the effect of MR blockade using canrenoate on both fast rate-sensitive and delayed feedback using an acute intravenous challenge of methylprednisolone, a steroid that displays both types of feedback on CORT secretion. Female Sprague-Dawley rats (14L:10D) were anesthetised (Hypnorm 0.15 ml i.m.; diazepam 0.75 mg i.p. per rat) and the jugular vein cannulated. Four days later cannulae were connected to an automated blood-sampling system (Windle et al. 1998). Blood samples were collected at 5-10 min intervals from 16:00h until 08:00h the following day (lights off 19:00h). One hour prior to lights off, intravenous injections of the following steroids were given: 5 mg potassium canrenoate in saline (CAN); 250 µg methylprednisolone sodium succinate (MethylPRED) or saline (VEH) 3 h prior to lights off (n=6-9 rats per group). CORT levels in blood were measured by RIA. Cort pulses were analysed using the PULSAR algorithm (Merriam & Wachter, 1982). Regarding the fast rate-sensitive feedback (18:00 -19:30h), compared to the VEH group, injections of MethylPRED resulted in decline in CORT concentrations that was significant (P<0.05; RM-ANOVA) by 37 min (83 ± 20; n=8 vs 25 ± 6; n=9; mean ± SEM). While CAN treatment alone resulted in a rapid (<25 min) elevation of CORT levels compared to VEH (212 ± 24; n=6 vs 74 ± 16; n=7; mean ± SEM), MethylPRED given 10 min after CAN produced CORT concentrations not different to CAN only treatment (RM-ANOVA). Thus blocking MR with CAN prevented the fast rate-sensitive effects of MethylPRED on CORT secretion. Regarding delayed feedback (19:30 – 0800h), following the MethylPRED challenge CORT levels remained low for 5 h until CORT pulsatility resumed at 22:48 ± 00:26h (mean ± SEM; n=9). Pre-treatment with CAN 10 min prior to MethylPRED significantly delayed (P<0.01 Kruskal-Wallis) the onset of CORT pulsatility for a further four and a half hours until 03:19 ± 01:13 h (mean ± SEM; n=6). During this time period treatment with CAN only did not suppress CORT pulsatility. Therefore blockade of MR with CAN enhanced delayed feedback of methylPRED on CORT secretion. In conclusion this study has shown that in vivo fast rate-sensitive feedback may be mediated through MRs. The observation that MR blockade enhanced delayed feedback was unexpected because similar work in our laboratory with the glucocorticoid receptor (GR) antagonist RU38486 blocked the delayed feedback of methylprednisolone (Andrews, 2001). One explanation is that delayed feedback via GR homodimers may be more effective than when heterodimers or MR homodimers are involved.