Search

The natural resistance of Mycobacterium abscessus to most commonly available antibiotics seriously limits chemotherapeutic treatment options, which is particularly challenging for cystic fibrosis patients infected with this rapid-growing mycobacterium. New drugs with novel molecular targets are urgently needed against this emerging pathogen. However, the discovery of such new chemotypes has not been appropriately performed. Here, we demonstrate the utility of a phenotypic screen for bactericidal compounds against M. abscessus using a library of compounds previously validated for activity against M. tuberculosis. We identified a new piperidinol-based molecule, PIPD1, exhibiting potent activity against clinical M. abscessus strains in vitro and in infected macrophages. Treatment of infected zebrafish with PIPD1 correlated with increased embryo survival and decreased bacterial burden. Whole genome analysis of M. abscessus strains resistant to PIPD1 identified several mutations in MAB_4508, encoding a protein homologous to MmpL3. Biochemical analyses demonstrated that while de novo mycolic acid synthesis was unaffected, PIPD1 strongly inhibited the transport of trehalose monomycolate, thereby abrogating mycolylation of arabinogalactan. Mapping the mutations conferring resistance to PIPD1 on a MAB_4508 tridimensional homology model defined a potential PIPD1-binding pocket. Our data emphasize a yet unexploited chemical structure class against M. abscessus infections with promising translational development possibilities.

Two new families of closely related selective, non-cytotoxic, and potent antitubercular agents were discovered: thioquinolines and thiazoloquinolines. The compounds were found to possess potent antitubercular properties in vitro, an activity that is dependent on experimental conditions of MIC determination (resazurin test and the presence or absence of Tween-80). To clarify the therapeutic potential of these compound families, a medicinal chemistry effort was undertaken to generate a lead-like structure that would enable murine efficacy studies and help elucidate the in vivo implications of the in vitro observations. Although the final compounds showed only limited levels of systemic exposure in mice, modest levels of efficacy in vivo at nontoxic doses were observed.

New chemical probes were synthesized to label galectin-3. They are based on the high affinity thiodigalactoside ligand. The probes were synthesized with benzophenone or acetophenone moieties as the photolabel for covalent attachment to the protein. Besides labeling the protein, these aromatic photolabels also greatly enhance the affinity of the probes towards galectin-3, due to the interaction of the photolabel with two arginine guanidinium groups of the protein. The linkage between the sugar and the photolabel was varied as an ester, an amide, and a triazole. For the amide and triazole derivatives, a versatile synthetic route towards a symmetrical 3-azido-3-deoxy-thiodigalactoside was developed. The new probes were evaluated for their binding affinity of human galectin-3. They were subsequently tested for their labeling efficiency, as well as specificity in the presence of a protein mixture and a human cancer cell lysate.

A rapid, efficient and versatile microwave-assisted dibutylstannylene-mediated 3-O-alkylation method for galactosides was established. The alkylation reaction was significantly enhanced by microwave irradiation and allowed the use of an alkylating agent too unstable for incorporation using conventional heating.