Mapping of Lysosomal Targeting Determinants of Cathepsin D

Abstract

The precursors of soluble mammalian lysosomal hydrolases are sorted from other, secretory glycoproteins as they exit the golgi apparatus, by virtue of binding to the mannose 6-phosphate glycoprotein receptor. This recycling receptor is responsible for the delivery to the lysosome of precursors carrying mannose 6-phosphate tags. It remains unclear how the cell distinguishes between lysosomally and non-lysosomally destined proteins, such that only the former are efficiently phosphorylated at the 6 position of mannose residues on their oligosaccharide side chains. Current evidence suggests that this discrimination is based on recognition of a stable domain in the folded structure of the lysosomal precursors and not of a primary sequence signal (Kornfeld, 1987). Cathepsin D is the only member of the aspartic protease gene family known to be lysosomal. Most of the other lysosomal cathepsins are thiol proteases. Cathepsin D shows a high degree of sequence identity to other aspartic proteases (Faust et al., 1985) and was therefore expected to have a 3-dimensional structure similar to that of pepsin, for which high resolution structures have been described. Thus, cathepsin D is a model of choice for identifying those residues which form the recognition domain necessary for phosphorylation and lysosomal targeting. To aid in prediction of the recognition domain, models of human, porcine and bovine cathepsins D were generated. Potentially important residues on cathepsin D were identified by comparison of the cathepsin D models to other aspartic protease structures. Residues conserved among the cathepsins D but not between them and renin and pepsin were changed by site-directed mutagenesis and the corresponding proteins expressed. The mutated cathepsins D were tested for the presence of phosphorylated mannose residues.