Structure and Topology of Presenilin Proteins curs within the domain encoded by exon 10 (Figure 1), The topology of PS-1 (Figure 1) was recently worked since the FAD-associated Δexon 10 mutation abolishes out by utilizing two strategies. First, putative TM dofragment formation, leading to the accumulation of fullmains were used to determine if they could export a length PS-1 (Thinakaran et al., 1996). Radiosequencing protease-sensitive substrate (Doan et al., 1996). Secof the N-terminus of the CTF revealed that the cleavage ond, cells permeabilized with the pore-forming toxin is heterogeneous and occurs after amino acids 291, 292, streptolysine O were probed with epitope-specific antiand 298 (Podlisny et al., 1997). Proteolytic processing bodies (Doan et al., 1996; DeStrooper et al., 1997). These of PS proteins appears to be a very slow process, and experiments suggest that PS-1 probably contains six or only minor amounts of the full-length PS proteins are eight TM domains, with both the N-and the C-terminal turned over into the respective fragments (Thinakaran domains, as well as the large hydrophilic loop between et al., 1996; Kim et al., 1997; Podlisny et al., 1997), indi-TM6 and TM7, oriented toward the cytoplasm (Figure cating that a second proteolytic mechanism, indepen-1). Li and Greenwald (1996) carried out very elegant experiments to determine the topology of the PS homo- dent of the cleavage within the domain encoded by exon 10, results in the rapid, efficient removal of PS proteins log in C. elegans (SEL-12). The membrane orientation of