To evaluate the effects of endurance training on gluconeogenesis and blood glucose homeostasis, trained as well as untrained short-term-fasted rats were injected with mercaptopicolinic acid (MPA), a gluconeogenic inhibitor, or the injection vehicle. Glucose kinetics were assessed by primed-continuous venous infusion of [U-14C]- and [6-3H]glucose at rest and during submaximal exercise at 13.4 m/min on level grade. Arterial blood was sampled for the determination of blood glucose and lactate concentrations and specific activities. In resting untrained sham-injected rats, blood glucose and lactate were 7.6 +/- 0.2 and 1.3 +/- 0.1 mM, respectively; glucose rate of appearance (Ra) was 71.1 +/- 12.1 mumol.kg-1.min-1. MPA treatment lowered blood glucose, raised lactate, and decreased glucose Ra. Trained animals had significantly higher glucose Ra at rest and during exercise. At rest, trained MPA-treated rats had lower blood glucose, higher blood lactate, and similar glucose Ra and disappearance rates (Rd) than trained sham-injected animals. Exercising sham-injected untrained animals had increased blood glucose and glucose Ra compared with rest. Exercising trained sham-injected rats had increased blood glucose and glucose Ra and Rd but no change in blood lactate compared with untrained sham-injected animals. In the trained animals during exercise, MPA treatment increased blood lactate and decreased blood glucose and glucose Ra and Rd. There was no measurable glucose recycling in trained or untrained MPA-treated animals either at rest or during submaximal exercise. There was no difference in running time to exhaustion between trained and untrained MPA-treated rats.(ABSTRACT TRUNCATED AT 250 WORDS)