Paul Mazur

Geranylgeranyltransferase I of Candida albicans: Null Mutants or Enzyme Inhibitors Produce Unexpected Phenotypes

Geranylgeranyltransferase I (GGTase I) catalyzes the transfer of a prenyl group from geranylgeranyl diphosphate to the carboxy-terminal cysteine of proteins with a motif referred to as a CaaX box (C, cysteine; a, usually aliphatic amino acid; X, usually L). The alpha and beta subunits of GGTase I from Saccharomyces cerevisiae are encoded by RAM2 and CDC43, respectively, and each is essential for viability. We are evaluating GGTase I as a potential target for antimycotic therapy of the related yeast, Candida albicans, which is the major human pathogen for disseminated fungal infections. Recently we cloned CaCDC43, the C. albicans homolog of S. cerevisiae CDC43. To study its role in C. albicans, both alleles were sequentially disrupted in strain CAI4. Null Cacdc43 mutants were viable despite the lack of detectable GGTase I activity but were morphologically abnormal. The subcellular distribution of two GGTase I substrates, Rho1p and Cdc42p, was shifted from the membranous fraction to the cytosolic fraction in the cdc43 mutants, and levels of these two proteins were elevated compared to those in the parent strain. Two compounds that are potent GGTase I inhibitors in vitro but that have poor antifungal activity, J-109,390 and L-269,289, caused similar changes in the distribution and quantity of the substrate. The lethality of an S. cerevisiae cdc43 mutant can be suppressed by simultaneous overexpression of RHO1 and CDC42 on high-copy-number plasmids (Y. Ohya et al., Mol. Biol. Cell 4:1017, 1991; C. A. Trueblood, Y. Ohya, and J. Rine, Mol. Cell. Biol. 13:4260, 1993). Prenylation presumably occurs by farnesyltransferase (FTase). We hypothesize that Cdc42p and Rho1p of C. albicans can be prenylated by FTase when GGTase I is absent or limiting and that elevation of these two substrates enables them to compete with FTase substrates for prenylation and thus allows sustained growth.

Selective glucocorticoid receptor nonsteroidal ligands completely antagonize the dexamethasone mediated induction of enzymes involved in gluconeogenesis and glutamine metabolism.

Glucocorticoids (GCs) are vital multi-faceted hormones with recognized effects on carbohydrate, protein and lipid metabolism. Previous studies with the steroid antagonist, RU486 have underscored the essential role of GCs in the regulation of these metabolic pathways. This article describes the discovery and characterization of novel GRalpha selective nonsteroidal antagonists (NSGCAs). NSGCAs 2 and 3 are spirocyclic dihydropyridine derivatives that selectively bind the GRalpha with IC(50s) of 2 and 1.5 nM, respectively. Importantly, these compounds are full antagonists of the induction by dexamethasone (Dex) of marker genes for glucose and glutamine metabolism; the tyrosine amino transferase (TAT) and glutamine synthetase (GS) enzymes, respectively. In contrast, GC-dependent transcriptional repression of the collagenase 1 (MMP-1) enzyme, an established GRalpha responsive proinflammatory gene; is poorly antagonized by these compounds. These NSGCAs might have useful applications as tools in metabolic research and drug discovery.

Discovery and optimization of novel 4-[(aminocarbonyl)amino]-N-[4-(2-aminoethyl)phenyl]benzenesulfonamide ghrelin receptor antagonists.

This Letter describes optimization of ghrelin receptor antagonists and inverse agonists starting from a screening hit.

The Structure of Neurospora crassa 3-Carboxy-cis,cis-Muconate Lactonizing Enzyme, a β Propeller Cycloisomerase

Muconate lactonizing enzymes (MLEs) convert cis,cis-muconates to muconolactones in microbes as part of the beta-ketoadipate pathway; some also dehalogenate muconate derivatives of xenobiotic haloaromatics. There are three different MLE classes unrelated by evolution. We present the X-ray structure of a eukaryotic MLE, Neurospora crassa 3-carboxy-cis,cis-muconate lactonizing enzyme (NcCMLE) at 2.5 A resolution, with a seven-bladed beta propeller fold. It is related neither to bacterial MLEs nor to other beta propeller enzymes, but is structurally similar to the G protein beta subunit. It reveals a novel metal-independent cycloisomerase motif unlike the bacterial metal cofactor MLEs. Together, the bacterial MLEs and NcCMLE structures comprise a striking structural example of functional convergence in enzymes for 1,2-addition-elimination of carboxylic acids. NcCMLE and bacterial MLEs may enhance the reaction rate differently: the former by electrophilic catalysis and the latter by electrostatic stabilization of the enolate.

Expression and characterization of protein geranylgeranyltransferase type I from the pathogenic yeast Candida albicans and identification of yeast selective enzyme inhibitors.

Protein geranylgeranyltransferase type I (GGTase I) is a heterodimeric zinc metalloenzyme catalyzing protein geranylgeranylation at cysteine residues present in C-terminal signature sequences referred to as CaaX (X=Leu) motifs. We have studied GGTase I as a potential antifungal target and recently reported its purification and cloning from the yeast Candida albicans (Ca GGTase I), an important human pathogen. Here, we report the high yield bacterial expression of Ca GGTase I by coexpression of maltose binding protein fusion proteins of both the alpha (Ram2p) and beta (Cdc43p) subunits. The cleaved and purified recombinant Ca GGTase I was demonstrated to be functional and structurally intact as judged by the presence of one equivalent of a tightly bound zinc atom and the near stoichiometric formation, isolation and catalytic turnover of a geranylgeranyl pyrophosphate-GGTase I complex. Kinetic analysis was performed with a native substrate protein, Candida Cdc42p, which exhibited significant pH dependent substrate inhibition, a feature not observed with other Ca GGTase I substrates. Prenyl acceptor substrate specificity was studied with a series of peptides in which both the CaaX motif, and the sequence preceding it, were varied. The prenyl acceptor K(M)s were found to vary nearly 100-fold, with biotinyl-TRERKKKKKCVIL, modeled after a presumably geranylgeranylated Candida protein, Crl1p (Rho4p), being the optimal substrate. A screen for inhibitors of Ca GGTase I identified compounds showing selectivity for the Candida versus human GGTase I. The most potent and selective compound, L-689230, had an IC(50) of 20 nM and >12,500-fold selectivity for Ca GGTase I. The lack of significant anti-Candida activity for any of these inhibitors is consistent with the recent finding that GGTase I is not required for C. albicans viability [R. Kelly et al., J. Bacteriol. 182 (2000) 704-713].

Purification of geranylgeranyltransferase I from Candida albicans and cloning of the CaRAM2 and CaCDC43 genes encoding its subunits.

All previously characterized protein geranylgeranyltransferases I (GGTase I) are heterodimeric zinc metalloenzymes which catalyse geranylgeranylation of a cysteine residue in proteins containing a C-terminal CaaL motif (C, Cys; a, aliphatic amino acid; L, Leu). The alpha and beta subunits of GGTase I of Saccharomyces cerevisiae are encoded by RAM2 and CDC43, respectively, and are essential for yeast viability. The authors are therefore investigating the role of geranylgeranylation in the related pathogenic yeast, Candida albicans, which is the most prevalent human fungal pathogen. GGTase I was purified to near homogeneity and also found to be a heterodimeric magnesium-dependent, zinc metalloenzyme displaying selectivity for CaaL-containing protein substrates. GGTase I peptide sequences were obtained from the purified protein and used to clone the genes encoding both subunits. CaRAM2 and CaCDC43 encode proteins that are 42 and 34% identical to their corresponding S. cerevisiae homologues, respectively, and 30% identical to their human homologues. Despite the limited overall homology, key zinc- and substrate-binding residues of the beta subunit (Cdc43p) are conserved. A unique feature of CaCdc43p is a tract of polyasparagine whose length varies from 6 to 17 residues among C. albicans strains and between alleles. Coexpression of both CaCDC43 and CaRAM2 under their native promoters complemented the ts defect of a S. cerevisiae cdc43 mutant but expression of the beta-subunit alone did not correct the growth defect, suggesting that hybrid GGTase I heterodimers are nonfunctional.

Discovery of betamethasone 17α-carbamates as dissociated glucocorticoid receptor modulators in the rat

A series of betamethasone 17alpha-carbamates were designed, synthesized, and evaluated for their ability to dissociate the two main functions of the glucocorticoid receptor, that is, transactivation and transrepression, in rat cell lines. A number of alkyl substituted betamethasone 17alpha-carbamates were identified with excellent affinity for the glucocorticoid receptor (e.g., 7, GR IC(50) 5.1 nM) and indicated dissociated profiles in functional assays of transactivation (rat tyrosine aminotransferase, TAT, and rat glutamine synthetase, GS) and transrepression (human A549 cells, MMP-1 assay). Gratifyingly, the in-vivo profile of these compounds, for example, 7, also indicated potent anti-inflammatory activity with impaired effects on glucose, insulin, triglycerides, and body weight. Taken together, these results indicate that dissociated glucocorticoid receptor modulators can be identified in rodents.