The role of corticosterone in human hypothalamic– pituitary–adrenal axis feedback

Abstract

Objective In humans, the glucocorticoid corticosterone circulates in blood at 10–20-fold lower levels than cortisol, but is found in higher relative amounts in postmortem brain samples. Access of cortisol and corticosterone to the central nervous system may not be equal. Additionally, the relative affinities for the glucocorticoid and mineralocorticoid receptors differ, such that corticosterone may play a significant role in human brain function. Design We measured cortisol and corticosterone levels in paired plasma and cerebrospinal fluid (CSF) samples. To test the relative potency of cortisol vs. corticosterone on hypothalamic–pituitary–adrenal (HPA) feedback, subjects underwent a three-phase, single-blind, randomized study assessing the postmetyrapone ACTH response over 3 h to an intravenous bolus of vehicle, cortisol or corticosterone (0·15 mg/kg and 0·04 mg/kg). Participants Outpatients undergoing diagnostic lumbar puncture who were subsequently deemed to be free of disease. Feedback was tested in healthy male volunteers. Measurements Plasma and CSF corticosterone to cortisol ratio was calculated and the ACTH response over time after the bolus glucocorticoid measured. Results Plasma corticosterone : cortisol was 0·069 ± 0·007; CSF corticosterone : cortisol was 0·387 ± 0·050 ( P < 0·001). Cortisol and corticosterone (0·15 mg/kg) suppressed ACTH vs. vehicle ( P = 0·002); there was no difference between corticosterone and cortisol. The 0·04 mg/kg dose had no effect on ACTH despite supraphysiological plasma corticosterone levels. Conclusions Corticosterone contributes almost 40% of total active glucocorticoids (cortisol and corticosterone) in the CSF. Significant effects on HPA axis suppression were only seen with supraphysiological levels of corticosterone, suggesting that corticosterone is not important in this model of nonstress-induced ACTH hypersecretion, in which the effect of cortisol predominates.