The distribution of cholesterol in the plasma membrane of epithelial cells has been determined using renal brush border vesicles as a model. In brush borders treated with Brevibacterium sp. or Nocardia erythropolis cholesterol oxidases, a significant fraction of the free cholesterol was oxidized rapidly, without glutaraldehyde fixation, and the remaining cholesterol was oxidized at a slower rate. The size of the readily accessible cholesterol pool, however, depended on the enzyme used, varying from 16% of the total in membranes treated with N. erythropolis oxidase, to 27% using the Brevibacterium sp. enzyme. The slowly accessible pool detected by the Brevibacterium oxidase was suppressed upon sphingomyelinase addition. On the other hand, the restricted activity of the Nocardia oxidase might depend on phosphatidylcholine/cholesterol interactions. These results indicate that cholesterol distribution is heterogeneous in intact renal brush border vesicles. They suggest that, as proposed for model system [Demel, R.A. Jansen, J.W.C.M., van Dijck, P.W.M., & van Deenen, L.L.M. (1977) Biochim. Biophys. Acta 465, 1-10], preferential interactions between some classes of phospholipids and cholesterol define cholesterol pools in the plasma membrane of epithelial cells.

Sphingomyelin (SM) and cholesterol are major lipid species of apical membranes in renal proximal tubular cells and confer to these membranes a low fluidity. Changes in membrane fluidity and/or lipidic composition were shown to affect the activity of cotransport systems of renal apical membranes. We evaluated the effect of decreasing membrane SM content on lipidic composition, membrane fluidity and sodium (Na)coupled uptakes in rabbit proximal tubular cells in primary culture. Sphingomyelinase (SMase) (30 to 250 mU/ml) decreased [3H]choline-labeled SM content, decreased cholesterol content, and increased cholesterol esterification. SMase did not modify membrane fluidity on isolated brush border membranes. SMase decreased Vmax of Na-dependent uptake of phosphate and alpha-methyl-D-glucoside, but not of alanine. SMase did not influence protein kinase C-induced inhibition of phosphate and glucose uptake. Increasing membrane cholesterol content with cholesterol-enriched liposomes subsequently to SMase action restored in part glucose uptake, but not phosphate uptake. In conclusion, SM degradation affected Na-phosphate and Na-glucose cotransports through changes in both SM and cholesterol contents of apical proximal membranes; these changes seemed to occur independently from changes in bulk membrane fluidity. These results suggest that SM and cholesterol have distinct and intricated roles in accessibility and/or activity of apical cotransport systems.

Treatment with cholesterol oxidases has shown that cholesterol is heterogeneously distributed in brush border membranes isolated from the apical domain of the renal and intestinal epithelial cells [Bloj, B., & Zilversmit, D. B. (1982) J. Biol. Chem. 257, 7608-7614; El Yandouzi, E. H., & Le Grimellec, C. (1992) Biochemistry 31, 547-551]. Cholesterol distribution between plasma membrane and intracellular membranes of the corresponding cells remains unexplored. The effects of Brevibacterium sp. cholesterol oxidase on the cholesterol content of LLC-PK1 cells, an epithelial cell line with multiple differentiated characteristics of the renal proximal tubule, were investigated. In confluent living cells grown as a monolayer on solid support, a small but significant fraction (13%) of the cholesterol was oxidized during the first hour of the oxidase treatment. Glutaraldehyde fixation prior to treatment resulted in a nearly complete (86.1 +/- 1.8) oxidation of the cellular cholesterol according to first-order kinetics. Filipin labeling and oxidation at 15 degrees C confirmed that cholesterol was essentially confined to the plasma membrane in LLC-PK1 cells. When adding the oxidase either on the apical or on the basolateral side of cells grown on permeant support and fixed with glutaraldehyde, a comparable monophasic oxidation of cholesterol was observed, despite the presence of efficient tight junctions. Adding the oxidase to both sides simultaneously did not increase the rate of oxidation. Finally, fixation of isolated renal brush border membranes with glutaraldehyde rendered undiscernible their cholesterol pools. We conclude that glutaraldehyde fixation, a commonly used process in the analysis of cholesterol distribution in cells, can mask the existence of cholesterol pools in plasma membranes.

No abstract