In vitro depalmitoylation of neurospecific peptides: implication for infantile neuronal ceroid lipofuscinosis.

Palmitoyl protein thioesterase 1 (PPT1) removes palmitate from specific cysteine residues in peptides and proteins. We have previously shown that a palmitoylated myelin glycoprotein. Po octapeptide (IRYCWLRR) can be specifically depalmitoylated by PPT1 in vitro (Cho and Dawson [1998] J. Neurochem. 171 ;323-329). To characterize further the substrate specificity of PPT1, we prepared various palmitoylated oligopeptides, based on palmitoylated sequences from different proteins. A truncated tetrapeptide from Po (RY[palmitoyl]-CW) was as good a substrate as the octapeptide Po, with optimal activity at pH 4.0. In contrast, other peptide substrates showed marked differences. Thus, the deacylation of GAP-43 (MLCCMRR), rhodopsin (VTTLCCGKN), and Galpha subunit (MGCLGNSK) peptides was more efficient at neutral pH (7.4) than at acidic pH (4.0), with the greatest efficiency toward the Galpha peptide (five- to sixfold higher than other substrates). Infantile neuronal ceroid lipofuscinosis (INCL) is caused by PPT1 deficiency, and the absence of enzymatic activity was confirmed with GAP-43 peptide as well as the Po peptide. LA-N-5 human neuroblastoma cells overexpressing PPT1 showed increased depalmitoylation of all the peptide substrates, indicating that these peptides are deacylated by PPT1. An amide derivative of a palmitoylated K-Ras peptide (AcG-palmitoyl diamino propionate-VKIKK) acted as an enzyme pseudosubstrate and inhibited PPT1 enzyme activity in a dose-dependent manner. The peptide itself (AcGCVKIKK) did not affect PPT activity. In summary, PPT1 is able to hydrolyze a range of cysteinyl peptide sequences found in both neuron-specific and ubiquitous (e.g., Galpha) proteins. The inhibitor of PPT1 activity should facilitate the development of a model for INCL and help explain the neuronal death in this disease.