PHARMACOKINETIC MODELING OF THE SINUSOIDAL EFFLUX OF ANIONIC LIGANDS FROM THE ISOLATED-PERFUSED RAT-LIVER - THE INFLUENCE OF ALBUMIN

This study contains a pharmacokinetic analysis on the efflux of organi c anions from the liver into the bloodstream (sinusoidal efflux) with specific reference to the influence of albumin. The net sinusoidal eff lux rate of dibromosulfophthalein (DBSP) from preloaded livers, being the resultant of sinusoidal efflux and reuptake of ligand by hepatocyt es downstream the sinusoid, can be strongly increased by the presence of bovine serum albumin (BSA), a protein having multiple binding sites for DBSP. We previously attributed this effect to a reduction of reup take through extracellular binding of the organic anion to the protein , lather than to an intrinsic stimulatory effect on the actual membran e transport process from tire cells. In the present study we tested th is hypothesis using a pharmacokinetic multicompartment liver model. Th is model resembles the parallel tube model in that the liver is descri bed by several compartments placed in series instead of a single well- stirred compartment and it takes into account rates of dissociation an d association in binding to proteins in the sinusoidal space. The mode l parameters were fitted from the sinusoidal efflux and biliary excret ion data from efflux experiments measuring the stimulatory effect of v arious concentrations of BSA. Equilibrium binding of DBSP to albumin a s well as the dissociation rate constant (k(off)) were determined in v itro with rapid filtration techniques. The experimental data could not be fitted satisfactorily when using the experimentally obtained value s of the protein association and dissociation rate constants (k(on) an d k(off)). However, they could be simulated accurately assuming 16 tim es higher values for the association and dissociation rate constant co mpared to those determined in vitro. Time constants of the perfusate f low, liver (re) uptake, and protein association nod dissociation indic ate that binding equilibrium does not exist,within the sinusoids and t hat, in particular at low protein concentrations, the net sinusoidal e fflux rate is association rate-limited: A large fraction of the ligand effluxed from the cell into the medium is taken up by the hepatocyte before binding to the proteins occurs. Higher k(on) and k(off) values predicted by the model might indicate altered DBSP-albumin binding cha racteristics upon passage through the liver but alternatively can be e xplained by an intrinsic effect of albumin on the carrier-mediated eff lux process. Efflux experiments showed a marked stimulatory effect of the protein on sinusoidal efflux but only a moderate effect on biliary excretion, despite a strong decrease in liver content. These patterns indicate that sinusoidal efflux and biliary excretion occur from two different intracellular compartments that equilibrate slowly.