Subjects (N=48) with poorly-controlled type 2 diabetes (HbA1c>7%) will be assigned to one of
the 4 following interventions in a randomised, parallel group design: [I] Control (placebo
injection; no exercise), [II] Anakinra (100 mg subcutaneous injection of human recombinant
interleukin-1 receptor antagonist), [III] Exercise (1 h cycle ergometry at 75% VO2max), [IV]
Anakinra + Exercise. Pancreatic beta-cell function (plasma insulin responses) will be
measured before and after each intervention using a hyperglycemic clamp (5.4 mM above basal
glucose) combined with GLP-1 infusion (0.5 pmol/kg/min) and arginine injection (5 g bolus).