Development of potent inhibitors of the enzyme DXS for the treatment of important infections such as malaria and tuberculosis
Isoprenoids are an extensive class of natural products that fulfil essential functions in all organisms and are biosynthesised from two common five-carbon precursors, isopentenyl diphosphate and its isomer dimethylallyl diphosphate. Whereas the methylerythritol phosphate (MEP) pathway for the precursor biosynthesis is used by most bacteria, algae, and plant chloroplasts, this pathway is absent in animals, making the enzymes of the MEP pathway attractive targets for the development of novel antibacterials, antimalarials or herbicides. In a systematic druggability assessment of all constituent enzymes of the MEP pathway based on their structures, we found that despite the generally hydrophilic character of the active sites, some subpockets are reasonably druggable and could therefore be addressed by structure- or fragment-based drug design. We focussed our attention on the first enzyme of the pathway, DXS, which catalyses the thiamine diphosphate-dependent decarboxylation and condensation of pyruvate and glyceraldehyde-3-phosphate.
In a de novo fragment-based drug design approach, we identified some promising fragments and confirmed their binding mode in solution using the STI NMR methodology. They occupy the thiamine-binding pocket and direct their exit vector towards the substrate-binding pocket. Owing to their structural features and inhibitory potency, they represent excellent thiamine analogue with reduced selectivity issues over other thiamine- or thiamine diphosphate-binding proteins.
In a separate approach, we focussed our attention on the identification of suitable thiamine-derived inhibitors of DXS, and identified a second scaffold, which directs its exit vector into the diphosphate-binding pocket.
Given the lack of structural information on pathogenic orthologues of DXS, we performed a ligand-based virtual screening campaign based on thiamine analogues as reference compounds, affording three hit classes with IC50 values in the low micromolar range and promising activity in cell-based assays with M. tuberculosis and P. falciparum DXS, which we currently optimise.
- Wirkstoffdesign und Optimierung - Prof. Dr. Hirsch / Prof. Dr. Hartmann