EVIDENCE THAT INTERFACIAL TRANSPORT IS RATE-LIMITING DURING PASSIVE CELL-MEMBRANE PERMEATION

The octanol to water overall transfer rate constants (k'(ow)), the octanol / water partition coefficients (K(pc)) and the diffusion coefficients in octanol (D(o)) and water (D(w)) were measured for 36 compounds. A plot of D(o)/k'(ow) as a function of D(o).K(pc)/D(w) was shown to fit a straight line for compounds having a high K(pc), and from the slope of this line, the thickness of the unstirred layer in the water phase was found to be 9.1-mu-m. Using this value and the data provided by compounds having a low K(pc), an estimate of 1.2-mu-m was obtained for the thickness of the unstirred layer in the octanol phase. This provided estimates for the true interfacial rate constants, k(ow) (for movement of a compound from octanol into water) and k(wo) (for movement in the opposite direction), which were corrected for the effects of the unstirred layers. k(ow) proved to be constant to within one order of magnitude for a series of compounds whose K(pc) values ranged over three orders of magnitude, while k(wo) for these same compounds varied directly as K(pc). Assuming that octanol has solvent properties similar to those of the lipid bilayer present in natural membranes, the permeability of these membranes to a given compound can be shown to be equal to k(wo)/2. Comparing the permeabilities calculated in this way to measured permeabilities of natural membranes showed them to be of the same order or smaller. These data are consistent with a proposal that the rate-limiting step in passive membrane permeation is not the rate of diffusion within the membrane itself, but rather, transfer of the permeant across the interfaces separating the lipid phase of the cell membrane from the aqueous phases on either side of it.