Selective anchoring in the specific plasma membrane domain: a role in epithelial cell polarity

If molecules anchored to the cytoskeleton are resistant to extraction by detergent, then we might expect the fraction of extractable molecules to correlate well with the fraction of molecules mobile in FRAP measurements on single cells. In fact, resistance to extraction correlates with the incidence of cells in which little or no recovery of fluorescence could be observed after photobleaching, the TIF of the population examined. This correlation is in fact likely to reflect the mobile fraction of labeled molecules on each cell. If the values for detergent extractability and, by extension, mobile fraction are normally distributed, then we expect a large fraction of cells to have a lower extractability than average. These cells would not show detectable mobility of labeled membrane proteins in a FRAP experiment. For example, 48 + 30% of B2-m is extractable from the basal domain of cells plated at confluency for 24 h. With this standard deviation, 16% of cells must have <18 % extractable, and if this correlates with mobility, these cells would certainly be scored in the TIE No cells had TIF for A2, which was completely extracted by TX-100. On the other hand, 30-52% of the cells showed TIF for B1 and B2-m in the basolateral membrane (where they are poorly extractable), while only 0-8 % showed TIF on the apical membrane (where their extractability is higher). Like the extraction experiments, the FRAP data are consistent with the existence of domain-specific interactions of basolateral antigens with the submembrane cytoskeleton, and mobile fractions present in both apical and basolateral domains.