Pseudo-Sugar Mimics of D-Glucosamine-6-phosphate are Activators of the <em>glmS</em> Ribozyme

Abstract

The glmS ribozyme is a gene-regulating riboswitch in bacteria whose enzymatic activity is dependent on D-glucosamine-6-phosphate (GlcN6P) as a natural cofactor. The GlmS enzyme plays a major role in the bacterial cell wall synthesis as it produces GlcN6P, an early intermediate in the biosynthesis of UDP-GlcNAc, which is an essential precursor of peptidoglycan. It is due to the presence of this ribozyme in human pathogens, including MRSA, Listeria monocytogenes, and Clostridium difficile that it represents an attractive target for the development of artificial activators. Carbohydrate mimics that possess significant structural dissimilarity to their natural counterparts, but efficiently mimic the activity of GlcN6P, could function as novel therapeutic entities.<br /> The substitution of the ring oxygen of natural sugars by functionalized methylene yields 5a-substituted carba-sugars. These promising synthetic glycomimetics could potentially interfere with their target structures in biological systems due to new functionalities. In the course of the project presented herein, the synthesis of fluoro-carba-sugar analogs of α-D-glucosamine and β-L-idosamine as well as a phenyl-carba-sugar variant of α-Dglucosamine was established. Titanocene(III)-catalyzed radical epoxide opening was used in the synthesis of (5aR)-fluoro-carba-α-D-glucosamine-6-phosphate as part of a new route towards base-labile carbocycles. The high diastereoselectivity of the hydrogen transfer step was rationalized by computation of the transition states involved in this step.<br /> In studies of the biological activity of the carba-sugar mimics on glmS ribozymes from different bacteria, fluoro-carba-GlcN6P effectively promoted the glmS ribozyme selfcleavage reaction, although with a significant loss of activity compared to its nonfluorinated analog. Docking of all synthesized carba-sugars to the rigid glmS binding pocket supported the experimental data, thereby allowing an in-depth structure-activity relationship. In contrast, the inhibiting effect of fluoro-carba-GlcN on B. subtilis growth, which is dependent on the GlcN-specific phosphoenolpyruvate:sugar phosphotransferase system, was comparable with that of carba-GlcN. The specific induction of cell envelope stress, thus, paves the way for a comprehensive analysis of the mode of action of fluoro-carba-GlcN.