Functional classification and validation of yeast prenylation motifs using machine learning and genetic reporters.

Protein prenylation by farnesyltransferase (FTase) is often described as the targeting of a cysteine-containing motif (CaaX) that is enriched for aliphatic amino acids at the a1 and a2 positions, while quite flexible at the X position. Prenylation prediction methods often rely on these features despite emerging evidence that FTase has broader target specificity than previously considered. Using a machine learning approach and training sets based on canonical (prenylated, proteolyzed, and carboxymethylated) and recently identified shunted motifs (prenylation only), this study aims to improve prenylation predictions with the goal of determining the full scope of prenylation potential among the 8000 possible Cxxx sequence combinations. Further, this study aims to subdivide the prenylated sequences as either shunted (i.e., uncleaved) or cleaved (i.e., canonical). Predictions were determined for Saccharomyces cerevisiae FTase and compared to results derived using currently available prenylation prediction methods. In silico predictions were further evaluated using in vivo methods coupled to two yeast reporters, the yeast mating pheromone a-factor and Hsp40 Ydj1p, that represent proteins with canonical and shunted CaaX motifs, respectively. Our machine learning-based approach expands the repertoire of predicted FTase targets and provides a framework for functional classification.

Protein Isoprenylation in Yeast Targets COOH-Terminal Sequences Not Adhering to the CaaX Consensus.

Protein isoprenylation targets a subset of COOH-terminal Cxxx tetrapeptide sequences that has been operationally defined as a CaaX motif. The specificity of the farnesyl transferase toward each of the possible 8000 combinations of Cxxx sequences, however, remains largely unresolved. In part, it has been difficult to consolidate results stemming from in vitro and in silico approaches that yield a wider array of prenylatable sequences relative to those known in vivo We have investigated whether this disconnect results from the multistep complexity of post-translational modification that occurs in vivo to CaaX proteins. For example, the Ras GTPases undergo isoprenylation followed by additional proteolysis and carboxymethylation events at the COOH-terminus. By contrast, Saccharomyces cerevisiae Hsp40 Ydj1p is isoprenylated but not subject to additional modification. In fact, additional modifications are detrimental to Ydj1p activity in vivo We have taken advantage of the properties of Ydj1p and a Ydj1p-dependent growth assay to identify sequences that permit Ydj1p isoprenylation in vivo while simultaneously selecting against nonprenylatable and more extensively modified sequences. The recovered sequences are largely nonoverlapping with those previously identified using an in vivo Ras-based yeast reporter. Moreover, most of the sequences are not readily predicted as isoprenylation targets by existing prediction algorithms. Our results reveal that the yeast CaaX-type prenyltransferases can utilize a range of sequence combinations that extend beyond the traditional constraints for CaaX proteins, which implies that more proteins may be isoprenylated than previously considered.