Final Activity Report Summary - SMARTSCREEN (Magnetic Resonance in Drug Screening) Recent developments in both hardware and software have enabled Nuclear magnetic resonance (NMR) to be a serious player in structure-based drug design. The partners involved in the SMARTSCREEN consortium identified a number of bottlenecks in this process and contributed with a number of methodological improvements in both the efficiency and the quality of drug discovery by means of NMR. It was the key objective that the previously acquired knowledge as well as the novel methods for the different stages of a drug discovery process is to be disseminated to the co-workers of the consortium. One of the most challenging tasks for both crystallography and NMR groups is the preparation of samples suitable for structure-based drug discovery. For the dissemination of knowledge activities as well as being relevant to the broad scientific community, the SMARTSCREEN consortium selected G protein-coupled receptors (GPCRs), Cytochromes and phosphates as general drug targets. Additionally, a steroid hormone enzyme and Matrix metalloproteinases (MMPs) were chosen due to the existing know-how made at former industrial cooperation. Screening of low complexity / low molecular weight compounds should identify novel scaffolds / fragments out of limited libraries of compounds with relatively low affinity. The resulting low affinity hits were to be structurally characterised and to be subsequently optimised to larger, higher affinity and specific inhibitors by means of chemical synthesis or the modification of functional key positions, respectively. Results for the target protein preparation: - The parallel expression strategy of multiple-construct proved particularly effective for structural analyses of multi-domain proteins by both NMR and X-ray crystallography. These proteins are composed of distinct structural modules and an optimal definition of the limits of these modules is a key parameter in attempts to obtain a 'well behaved' sample suitable for structure determination. - A global analysis of the expression strategies and the results gathered shows that expression was mainly performed in Escherichia coli based pipelines which includes strains allowing the co-expression of rare tRNAs with the protein of interest have been widely used and for the difficult cases, eukaryotic expression systems such as baculovirus or mammalian cells were necessary. - Additionally, efforts spent on targets from families of closely related proteins exploiting standardised refolding protocols contributed to the success rate and the use of Mass spectrometry (MS) assisted the effective sample preparation. Results for the screening endeavour to identify novel drug candidates: - For the screening, ligand-based rather than protein-based detection methods were applied since this approach (i) does not require isotope labelled and assigned proteins, (ii) requires much less instrument time, (iii) consumes significantly less protein per data point. For the detected binders, the binding sites of the ligand to the target protein were detected by NMR using protein-based 2D-methods. For projects for which no labelled and assigned proteins were available, the binding sites were determined with a novel ligand-based 1D NMR competition assay. Interesting drug candidates which appeared to be easily chemically modifiable were subsequently submitted to the collection of structural restraints for NMR-based complex models or to the structure determination by crystallography. Results for the bio-structures of the protein-lead compound complexes: - Steering medicinal chemists through the lead optimisation process by detailed information about protein-ligand interactions led to remarkable success in the development of novel drugs. Applicable tools were developed which ease the requirements for NMR data acquisition to the calculation of docking complexes with small molecules, which help medicinal chemists to judge structure activity relationships. - The advances in modelling and NMR data-driven docking procedures allow the determination of approximate structures of biomolecular complexes with rising precision.
