Mycobacterial arabinan biosynthesis: Characterization of AftB and AftC arabinosyltransferase activity using synthetic substrates
Tuberculosis (TB) is a chronic infectious disease caused by M. tuberculosis (Mtb). Treatment of TB is prolonged, and multidrug resistant (MDR-TB) and extensively drug resistant TB (XDR-TB) cases are ever increasing. Efforts to discover and develop new drugs have increased in recent years so improvement of the existing therapies is urgently needed. The cell wall of Mtb with its unique physiological properties has historically been an important and valid drug target. Mycobacterial arabinosyltransferases are membrane bound glycosyltransferases involved in the biosynthesis of the arabinan portion of two major polysaccharides, arabinogalactan (AG) and lipoarabinomannan (LAM), associated with the cell wall. In this work, M. smegmatis was used as a model organism to study arabinosyltransferases and the biosynthesis of cell wall arabinofuran—the main constituent of AG and LAM.
This dissertation addresses the development of a cell free arabinosyltransferase assay for AftB and AftC glycosyltransferases. Since it was not possible to express AftB transmembrane protein, we probed AftB transferase activity from the crude membranes using a synthetic arabinose disaccharide acceptor. In this study, a robust cell free radioactive arabinosyltransferase assay was developed using a linkage specific synthetic disaccharide acceptor. Relative mobility of the enzymatic product on a thin layer chromatogram and autoradiography clearly demonstrated the enzymatic conversion of the disaccharide acceptor to a trisaccharide. The trisaccharide product was further confirmed by matrix assisted laser desorption ionization- time of flight (MALDI-TOF) or high pH anion exchange chromatography (HPAEC) analysis. GC-MS analysis showed that the additional arabinose added to the enzymatic product was (1→2) linked. This assay is dependent on time and enzyme concentration and the product formation is not sensitive to the action of ethambutol or the absence of the putative arabinosyltransferases encoded by embA, embB or embC. Additionally, further optimal conditions were determined including buffers, pH, divalent cations, detergents, and the effect of alkylating and reducing agents.
In a contiguous study, we wanted to convert this assay into a non-radioactive assay for the screening of compounds. We successfully developed an ELISA based non-radiolabeled arabinosyltransferase assay using CS35-mAb that recognized (1→2) linked enzymatic product. As this method involves several washing steps and time consuming processes, we further modified the substrate with fluorescent labeling, however, we failed to establish a fluorescence based arabinosyltransferase assay.
Unlike AftB, in a parallel study we were successful in expressing, solubilizing and purifying Mtb AftC. We developed an in vitro transferase assay using purified recombinant AftC and demonstrated that AftC retains transferase activity only when reconstituted into proteoliposomes. Additionally, we were successful in synthesizing alternate arabinose donors Z-nerylphosphoryl D-arabinose (Z-NPA), Z,Z-farnesylphosphoryl D-arabinose (Z-FPA), E,E-farnesylphosphoryl D-arabinose (E-FPA), and Z,Z,Z,Z,E,E-heptaprenylphosphoryl D-arabinose (Z-HPA). Among these lipid donors, Z-FPA demonstrated better solubility in the assay buffer compared to the native donor decaprenyphophoryl D-arabinose (DPA).
This dissertation describes screening of drug candidate compounds using a microtiter plate based whole cell Alamar Blue assay. We have successfully screened nearly three thousand compounds for Mtb growth inhibition.