Considerations in the design of hyperinsulinemic-euglycemic clamps in the conscious mouse

Abstract

Despite increased use of the hyperinsulinemic-euglycemic clamp to study insulin action in mice, the effects of experimental parameters on the results obtained have not been addressed. In our studies, we determined the influences of sampling sites, fasting duration, and insulin delivery on results obtained from clamps in conscious mice. Carotid artery and jugular vein catheters were implanted in C57BL/6J mice (n = 6-10/group) fed a normal diet for sampling and infusions. After a 5-day recovery period, mice underwent a 120-min clamp (2.5-mU . kg(-1) . min(-1) insulin infusion; approximately 120-130 mg/dl glucose) while receiving [3-(3)H]glucose to determine glucose appearance (endoR(a)) and disappearance (R(d)). Sampling large volumes (approximately 100 mul) from the cut tail resulted in elevated catecholamines and basal glucose compared with artery sampling. Catecholamines were not elevated when taking small samples ( approximately 5 mul) from the cut tail. Overnight (18-h) fasting resulted in greater loss of total body, lean, and fat masses and hepatic glycogen but resulted in enhanced insulin sensitivity compared with 5-h fasting. Compared with a 16-mU/kg insulin prime, a 300-mU/kg prime resulted in hepatic insulin resistance and slower acquisition of steady-state glucose infusion rates (GIR) after a 5-h fast. The steady-state GIR was expedited after the 300-mU/kg prime in 18-h-fasted mice. The GIR and R(d) rose with increasing insulin infusions (0.8, 2.5, 4, and 20 mU . kg(-1) . min(-1)), but endoR(a) was fully suppressed with doses higher than 0.8 mU . kg(-1) . min(-1). Thus, common variations in experimental factors yield different results and should be considered in designing and interpreting clamps.