Biochemical Elucidation of the Isoprenoid Binding Site of the Yeast Isoprenylcysteine Carboxyl Methyltransferase, Ste14p

CaaX proteins undergo three post-translational modifications: isoprenylation of the cysteine residue, endoproteolysis of the –aaX residues and methylation of the isoprenylated cysteine by Isoprenylcysteine carboxyl methyltransferase (Icmt) (1). The most notable of approximately 300 CaaX proteins are the Ras superfamily of proteins along with various others such as Rheb proteins, nuclear lamins and the yeast a-factor mating pheromone (2). Specifically, Ras proteins are small GTPases that are involved in cell growth, differentiation and survival. Mutations in the KRAS isoform are associated with approximately 30% of all cancers and 90% of pancreatic cancers (3–5). Past research revealed Icmt as a vital enzyme for proper localization of K-Ras and oncogenic transformation (6–8). Also, given the vast number of cellular processes regulated by CaaX processing, it is essential to elucidate the structure and function of Icmt. Currently, there is limited structural information for Icmt, an integral membrane protein localized in the endoplasmic reticulum (9, 10). We are focusing on an in-depth analysis of the structure and function of Ste14p, the Icmt from Saccharomyces cerevisiae, as a model for the human enzyme. As methylation is essential for the proper localization and oncogenic transformation of K-Ras, Icmt may prove to be an excellent chemotherapeutic target. Ste14p is an integral membrane protein and as a result, research has been limited on ways to identify the isoprenoid binding site. We designed photoreactive analogs based on minimal substrates, N-acetyl-S-farnesyl-L-cysteine (AFC) and the yeast a-factor peptide mating pheromone, of Ste14p that contain benzophenone or diazirine moieties in either the lipid or peptide portion of the molecule (11, 12). We optimized and determined that these analogs were substrates of His-Ste14p and labeled efficiently for future applications. Also, the use of a diazirine photoactive moiety allowed for more efficient labeling and revealed additional methods for crosslinked peptide identification (12). To expand the tools needed to fully understand the substrate binding site of Ste14p, our lab aligned 15 species of Icmt and determined 77 conserved residues (13). We generated a large library of His-Ste14p mutants through alanine scanning and extended mutagenesis of the conserved residues. These mutants were tested for substrate specificity with two known substrates of Ste14p. Wild-type Ste14p has been shown to not prefer either substrate (14). Several residues within transmembrane regions 1 and 2 were identified to vary in substrate specificity from wild-type. The most significant was residue L56 which appeared to severely decrease recognition of N-acetyl-S -farnesyl-L-cysteine (AFC) while retaining activity with N-acetyl-S-geranylgeranyl-L-cysteine (AGGC). These results narrowed in on areas of interest within His-Ste14p that may be a part of the substrate binding site. Additionally, we designed and purified single cysteine mutants in the cys-less background of His-Ste14p. We were then able to use the photoreactive substrate analogs and the purified cysteine mutants to understand in detail where Ste14p accommodates the lipophilic isoprenylated protein substrate. Using the cysteine-specific cleavage reagent 2-nitro-5-thiocyanobenzoic acid (NTCB) and various His-Ste14p cysteine mutants, we demonstrated that the AFC photoreactive analogs containing a benzophenone moiety in the amide or farnesyl moiety photolabeled between residues 47-77 of the protein. Additionally, we visualized the labeled peptide fragments via Coomassie stained gel which indicated substantial protein available for subsequent identification using mass spectrometry. This will aid in determining the exact amino acid sequence participating in substrate recognition within Ste14p. Once this knowledge is identified, the development of more potent inhibitors of Icmt will be synthesized for K-Ras driven cancers. Finally, while understanding the biochemical aspects of Icmt, we also developed a readily-synthesized library of human Icmt (hIcmt) inhibitors based on the minimal substrate N-acetyl-S-farnesyl-L-cysteine (AFC). Elaboration of the structure-function relationships of these compounds using an in vitro methyltransferase assay led to our current lead compound, STAB-F3-Diol, which demonstrated a mostly competitive inhibition with a Ki of approximately 245 nM against hIcmt. Further optimization of the biochemical activity and drug-like characteristics of this potent and promising agent resulted in the development of a new compound, STAB-F3-Alcohol. This compound also exhibited a mostly competitive inhibition with a K i value of 75 nM against hIcmt in our in vitro enzymatic assay. We are currently testing this molecule in various cellular assays to determine the effect of this compound on mutant K-Ras driven cancer cells. Together, these data will be used to interpret Icmt structure, function and catalysis of oncogenic CaaX proteins such as K-Ras as a means of designing potent chemotherapeutic inhibitors of Icmt