Abstract

Trichloroacetic acid (TCA), a mouse liver carcinogen, is a drinking water contaminant and a metabolite of solvents such as trichloroethylene and perchloroethylene. Because acidic drugs are often bound more strongly to human than to rodent plasma proteins, a study was undertaken to determine whether this was the case for TCA and to clarify the mechanistic bases for species differences. Equilibrium dialysis was used to measure in vitro binding of a range of TCA concentrations to plasma of humans, rats, and mice. Plots of observed data for free versus bound TCA concentrations were compared with simulations from each of three binding models: a single saturable site model; a saturable plus nonsaturable site model; and a two-saturable site model. Dissociation values (K_d) did not differ significantly from one species to another, but N (number of binding sites/molecule) ranged from 2.97 for humans to 0.17 for mice. Binding capacities (B_max) for humans, rats, and mice were 709, 283, and 29 μM, respectively. The greater plasma protein binding of TCA in humans would be expected to not only increase the residence time of the compound in the bloodstream, but to substantially reduce the proportion of TCA that is available for uptake by the liver and other tissues. Species differences in the bound fraction diminished at very low, environmentally relevant TCA concentrations, but the percentage bound increased markedly. These findings suggest that the practice of using total blood levels of TCA as a dose metric in interspecies extrapolation of cancer risks needs to be re-examined.