Human plasma contains cortisol (F) and corticosterone (B) at a ratio of ~10:1. B is well studied in mice and rats, which do not produce F due to absent adrenal Cyp17, but is largely neglected in humans. Differential transmembrane export of F > B by ABCB1 may account for accumulation of B in the CNS. Conversely, ABCC1, expressed in human adipose tissue, preferentially exports B>F. Here we tested the hypotheses that: (i) negative feedback suppression of the hypothalamic-pituitary-adrenal (HPA) axis is disproportionately sensitive to B; (ii) adipose tissue is disproportionately sensitive to F; and (iii) low plasma B contributes to impaired HPA axis negative feedback and increased F action in metabolic syndrome. We validated a stable isotope tracer for B in vitro and demonstrated distinct kinetics of B and F in vivo. In a randomised crossover study, we undertook ramped steady state infusion of B or F in 10 patients with Addison’s disease. Although levels of B were marginally lower than F, ACTH was similarly suppressed, and yet glucocorticoid-responsive transcripts in adipose tissue were much higher following F than B (PER1 2.2-fold and LPL 1.3-fold; p < 0.05). We assessed associations of ACTH-stimulated plasma B and F with features of metabolic syndrome in a cross-sectional study (n=279). Glucose tolerance was impaired with higher F (β=0.146, p=0.01) but lower B (β = -0.056, p = 0.05). These data support the concept of differential tissue sensitivity to B and F, whereby B suppresses the HPA axis more effectively than it induces adverse effects in adipose tissue. Enhanced CYP17 activity, causing ‘relative B deficiency’, may contribute to HPA axis activation and enhanced F action in adipose tissue in obesity. B therapy might allow control of HPA axis activation without inducing adverse metabolic effects. The ‘neglected second glucocorticoid’, corticosterone, may optimise glucocorticoid action in the human CNS, and simultaneously limit adverse metabolic effects driven by cortisol excess.