Unprecedently Large-Scale Kinase Inhibitor Set Enabling the Accurate Prediction of Compound-Kinase Activities: A Way toward Selective Promiscuity by Design?

Drug discovery programs frequently target members of the human kinome and try to identify small molecule protein kinase inhibitors, primarily for cancer treatment, additional indications being increasingly investigated. One of the challenges is controlling the inhibitors degree of selectivity, assessed by in vitro profiling against panels of protein kinases. We manually extracted, compiled, and standardized such profiles published in the literature: we collected 356 908 data points corresponding to 482 protein kinases, 2106 inhibitors, and 661 patents. We then analyzed this data set in terms of kinome coverage, results reproducibility, popularity, and degree of selectivity of both kinases and inhibitors. We used the data set to create robust proteochemometric models capable of predicting kinase activity (the ligand-target space was modeled with an externally validated RMSE of 0.41 ± 0.02 log units and R02 0.74 ± 0.03), in order to account for missing or unreliable measurements. The influence on the prediction quality of parameters such as number of measurements, Murcko scaffold frequency or inhibitor type was assessed. Interpretation of the models enabled to highlight inhibitors and kinases properties correlated with higher affinities, and an analysis in the context of kinases crystal structures was performed. Overall, the models quality allows the accurate prediction of kinase-inhibitor activities and their structural interpretation, thus paving the way for the rational design of compounds with a targeted selectivity profile.

Structure-based design of 7-azaindole-pyrrolidine amides as inhibitors of 11ß-hydroxysteroid dehydrogenase type I.

Indole-pyrrolidines were identified as inhibitors of 11ß-hydroxysteroid dehydrogenase type 1 (11ß-HSD1) by high-throughput screening. Optimisation of the initial hit through structure-based design led to 7-azaindole-derivatives, with the best analogues displaying single digit nanomolar IC(50) potency. The modeling hypotheses were confirmed by solving the X-ray co-crystal structure of one of the lead compounds. These compounds were selective against 11ß-hydroxysteroid dehydrogenase type 2 (selectivity ratio >200) and exhibited good inhibition of 11ß-HSD1 (IC(50)<1µM) in a cellular model (3T3L1 adipocytes).

Discovery and structure-activity relationships of pentanedioic acid diamides as potent inhibitors of 11beta-hydroxysteroid dehydrogenase type I.

Benzylamides of pentanedioic acid were identified as inhibitors of 11beta-hydroxysteroid dehydrogenase type 1 (11beta-HSD1) by high-throughput screening. Optimisation to 2-adamantyl amides yielded inhibitors with single digit nanomolar IC(50)s on the 11beta-HSD1 human isoform. The hydroxy adamantyl amide lead compound was selective against 11beta-hydroxysteroid dehydrogenase type 2 (selectivity ratio >1000) and displayed good inhibition of 11beta-HSD1 (IC(50)<0.1microM) in a cellular model (3T3L1 adipocytes).

Discovery and structure-guided drug design of inhibitors of 11beta-hydroxysteroid-dehydrogenase type I based on a spiro-carboxamide scaffold.

Spiro-carboxamides were identified as inhibitors of 11beta-hydroxysteroid-dehydrogenase type 1 by high-throughput screening. Structure-based drug design was used to optimise the initial hit yielding a sub-nanomolar IC(50) inhibitor (0.5nM) on human 11beta-HSD1 with a high binding efficiency index (BEI of 32.7) which was selective against human 11beta-HSD2 (selectivity ratio>200000).

Molecular modelling studies on the interactions of human DNA topoisomerase IB with pyridoxal-compounds.

Candida guilliermondii and human DNA topoisomerases I are inhibited by PL (pyridoxal), PLP (pyridoxal 5'-phosphate) and PLP-AMP (pyridoxal 5'-diphospho-5'-adenosine) (PL

Pyridoxal 5'-phosphate inactivates DNA topoisomerase IB by modifying the lysine general acid.

The present results demonstrate that pyridoxal, pyridoxal 5'-phosphate (PLP) and pyridoxal 5'-diphospho-5'-adenosine (PLP-AMP) inhibit Candida guilliermondii and human DNA topoisomerases I in forming an aldimine with the epsilon-amino group of an active site lysine. PLP acts as a competitive inhibitor of C.guilliermondii topoisomerase I (K(i) = 40 microM) that blocks the cleavable complex formation. Chemical reduction of PLP-treated enzyme reveals incorporation of 1 mol of PLP per mol of protein. The limited trypsic proteolysis releases a 17 residue peptide bearing a lysine-bound PLP (KPPNTVIFDFLGK*DSIR). Targeted lysine (K*) in C.guilliermondii topoisomerase I corresponds to that found in topoisomerase I of Homo sapiens (K532), Candida albicans (K468), Saccharomyces cerevisiae (K458) and Schizosaccharomyces pombe (K505). In the human enzyme, K532, belonging to the active site acts as a general acid catalyst and is therefore essential for activity. The spatial orientation of K532-PLP within the active site was approached by molecular modeling using available crystallographic data. The PLP moiety was found at close proximity of several active residues. PLP could be involved in the cellular control of topoisomerases IB. It constitutes an efficient tool to explore topoisomerase IB dynamics during catalysis and is also a lead for new drugs that trap the lysine general acid.

Structure-based virtual screening: an application to human topoisomerase II alpha.

The eukaryotic topoisomerase II is involved in several vital processes, such as replication, transcription, and recombination. Many compounds interfering with the catalytic action of this enzyme are efficient in human cancer chemotherapy. We applied a methodology combining molecular modeling and virtual screening techniques to identify human topoisomerase II alphainhibitors. Data from structural biology and enzymatic assays together with a good background on the enzyme mechanism of action were helpful in the approach. A human topoisomerase II alpha model provided an insight into the structural features responsible for the activity of the enzyme. A protocol comprising several substructural and protein structure-based three-dimensional pharmacophore filters enabled the successful retrieving of inhibitors of the enzyme from large databases of compounds, thus validating the approach. A subset of protein structural features required for the enzyme inhibition at the protein-DNA interface were identified and incorporated into the pharmacophore models. Compounds sharing a DNA-intercalating chromophore and a moiety interfering with the protein active site emerged as good inhibitors.