Protein Modifications and Their Significance in rab5 Function

Abstract

Rab5 belongs to a subset of proteins that are post-translationally modified at their carboxytermini via geranylgeranylation and carboxymethylation. We are studying hydrophobic modifications of cell-free synthesized rab5 and how these structural additions contribute to rab5 function. Prenylation of nascent peptide can be identified by in vitro translation in the presence of the appropriate radiolabelled precursor. We find that cell-free synthesized rab5 exists in at least two isoforms, with the addition of mevalonate (100 μM) shifting the peptide to a greater mobility on urea-gradient gels (Fig. 1). The isoform of greater mobility incorporates radiolabel upon addition of [3H]mevalonate to the in vitro system, clearly demonstrating that this form of the peptide is polyisoprenylated. Cell-free synthesized rab5 will also incorporate radiolabel in the presence of [3H]S-adenosyl methionine, however, we have found this reaction to be substoichiometric and non-specific. Additional factors appear to be required for the methylesterification of rab5 in vitro. In contrast, expression of rab5 in the cell-free reaction strongly enhances the stable methylation of an endogenous protein that runs with a broad mobility of 17/19kD (Fig. 2). The novel 17/19kD species is also methylated in response to the presence of proteins implicated to participate in vesicle traffic, including βγ subunits of trimeric G proteins. While the alpha subunit of Gi₃ also supports modification of p17/19kD, related factors (e.g. alpha subunits of Gi₁ and Gs) do not. The pattern of methylation in response to various GTP-binding proteins indicates that elements specifically involved in vesicle traffic interact with p17/19 to influence its methylesterification. What role this may play, for example, in rab5 endocytic function is still unclear. To identify domains of interaction, a series of truncated rab5 molecules have been constructed and expressed in vitro. Our analysis of the functional ability to support methylation of p17/19 indicates that while rab5 molecules lacking C-terminal cysteines, and thus non-prenylated forms, support methylesterification of p17/19, deletion of 16 upstream amino acids produces a form of rab5 that is apparently unable to influence the methylation of p17/19. Thus, rab5 C-terminal structural elements, exclusive of the isoprenylated cysteine residues, are required for interaction with p17/19. We are continuing studies to define the relative contributions of hydrophobic modifications of both rab5 and p17/19 in endocytic vesicle traffic.