Futile cycling between 4-methylumbelliferone and its sulfate and glucuronide conjugates was examined in the single-pass perfused rat liver preparation. The steady-state hepatic extraction ratio of 4-methylumbelliferone was found to be high (0.97) at a low input concentration of 0.005 mumol/L (tracer), with a net 4-methylumbelliferyl sulfate/4-methylumbelliferyl glucuronide ratio of about 5:1; at 63 mumol/L the steady-state extraction ratio had remained constant despite a shift from net sulfation to net glucuronidation. At higher input 4-methylumbelliferone concentrations, saturation was evidenced by a decreased steady-state extraction ratio and reduced net sulfation and net glucuronidation. Because 4-methylumbelliferyl sulfate and 4-methylumbelliferyl glucuronide deconjugation would result in an intracellular accumulation of 4-methylumbelliferone, the phenomenon was monitored with a shift in tracer [3H]4-methylumbelliferone metabolism from sulfation to glucuronidation with increased intracellular 4-methylumbelliferone concentration. When 4-methylumbelliferyl sulfate (0 to 890 mumol/L) or 4-methylumbelliferyl glucuronide (0 to 460 mumol/L) was delivered simultaneously with tracer [3H]4-methylumbelliferone to the rat liver, notable desulfation of 4-methylumbelliferyl sulfate (18% to 38% rate in) but little deglucuronidation of 4-methylumbelliferyl glucuronide (1.2% to 2.1% rate in) was observed. With 4-methylumbelliferyl sulfate, 4-methylumbelliferone and 4-methylumbelliferyl glucuronide were readily found as metabolites, whereas with 4-methylumbelliferyl glucuronide, levels of the metabolites, 4-methylumbelliferone and 4-methylumbelliferyl sulfate, were much reduced. 4-Methylumbelliferyl sulfate and not 4-methylumbelliferyl glucuronide shifted tracer [3H]4-methylumbelliferone metabolism from [3H]4-methylumbelliferyl sulfate to [3H]4-methylumbelliferyl glucuronide formation in a concentration-dependent fashion. The steady-state extraction ratio for 4-methylumbelliferyl sulfate (0.1 to 0.3) was comparatively higher than that for 4-methylumbelliferyl glucuronide (0.05), and it was found to increase with concentration, an observation explained by the nonlinear protein binding of 4-methylumbelliferyl sulfate. Biliary excretion rates for 4-methylumbelliferone and 4-methylumbelliferyl sulfate were proportional to their input or net formation rates, regardless of whether 4-methylumbelliferone, 4-methylumbelliferyl glucuronide or 4-methylumbelliferyl sulfate was administered. By contrast, the excretion rate of 4-methylumbelliferyl glucuronide when administered was only 1/25 the excretion of 4-methylumbelliferyl glucuronide formed from 4-methylumbelliferone and 4-methylumbelliferyl sulfate. The extent of choleresis paralleled the excretion patterns of preformed and formed 4-methylumbelliferyl glucuronide; bile flow was normal with 4-methylumbelliferyl glucuronide administration and was markedly enhanced with increased 4-methylumbelliferone or 4-methylumbelliferyl sulfate administration. The data suggest the presence of a transmembrane barrier for entry of 4-methylumbelliferyl glucuronide and not 4-methylumbelliferyl sulfate or 4-methylumbelliferone into hepatocytes.(ABSTRACT TRUNCATED AT 400 WORDS)