Non-electrolyte solute permeabilities of human placental microvillous and basal membranes.

1. Permeability to non-electrolytes of isolated microvillous and basal membranes from human term placenta was measured using stopped-flow light-scattering techniques. The studied solutes were urea, ethylene glycol, glycerol, creatinine, erythritol, arabitol and mannitol. 2. At 37 degrees C, permeability of the microvillous membrane to mannitol and urea was 0.30 +/- 0.02 x 10(-6) cm/s (mean +/- S.E.M.) and 3.2 +/- 0.2 x 10(-6) cm/s, respectively. The corresponding permeabilities for the basal membrane were 1.2 +/- 0.1 x 10(-6) cm/s (mannitol) and 4.4 +/- 0.3 x 10(-6) cm/s (urea). The basal membrane was substantially more permeable to hydrophilic solutes than the microvillous membrane. This is probably due to differences in lipid composition, as illustrated by membrane cholesterol content, which was found to be approximately 50% lower in the basal as compared to the microvillous membrane. 3. Similarities between permeabilities in placental membranes and lipid bilayers and the linear relationship noted between solute hydrophobicity and placental permeability suggested that solutes permeate both human syncytiotrophoblast membranes by a solubility/diffusion mechanism. In the microvillous membrane this was supported by data obtained for activation energies (> 10 kcal/mol) and reflection coefficients (close to 1). In the basal membrane, low activation energies for glycerol and urea and a low reflection coefficient for urea indicated that these solutes may, in part, share a common pathway with water. 4. It was estimated that the placental permeability to molecules with a molecular weight under 200 observed in vivo can, to a great extent, be accounted for by transcellular permeation.