The effects of endurance training (running 90 min/day at 30 m/min, 10% grade) on hepatic gluconeogenesis were studied in 24-h-fasted rats with use of the isolated liver perfusion technique. After isolation, the liver was perfused (single pass) for 30 min with Krebs-Henseleit bicarbonate buffer and fresh bovine erythrocytes (hematocrit 22-24%) with no added substrate. Subsequent to the "washout" period, the reservoir was elevated with various concentrations of lactate and [U-14C]lactate (10,000 dpm/ml) to assess hepatic glucose production. Relative flow rates were not significantly different between trained (1.94 +/- 0.05 ml/g liver) and control livers (1.91 +/- 0.05 ml/g liver). Furthermore, no significant differences were observed in perfusate pH, hematocrit, bile production, or serum alanine aminotransferase effluxing from trained or control livers. At saturating arterial lactate concentrations (> 2 mM), the maximal rate (Vmax) for hepatic glucose production was significantly higher for trained (0.91 +/- 0.04 mumol.min-1 x g liver-1) than for control livers (0.73 +/- 0.02 mumol.min-1 x g liver-1). That this reflected increased gluconeogenesis is supported by a significant elevation in the Vmax for [14C]glucose production from trained (13,150 +/- 578 dpm.min-1 x g liver-1) compared with control livers (10,712 +/- 505 dpm.min-1 x g liver-1). Significant increases were also observed in the Vmax for lactate uptake (25%), O2 consumption (19%), and 14CO2 production (23%) from endurance-trained livers. The Km for hepatic glucose output, approximately 1.05 mM lactate, was unchanged after endurance training. These findings demonstrate that chronic physical activity results in an elevated capacity for hepatic gluconeogenesis, as assessed in situ at saturating lactate concentrations.