Andreas H. Göller

CypScore: Quantitative prediction of reactivity toward cytochromes P450 based on semiempirical molecular orbital theory

CypScore predicts the reactivity of competing positions in the same and different molecules to a variety of cytochrome P450 metabolic reactions on a single reactivity scale.

Best of both worlds: combining pharma data and state of the art modeling technology to improve in Silico pKa prediction.

In a unique collaboration between a software company and a pharmaceutical company, we were able to develop a new in silico pKa prediction tool with outstanding prediction quality. An existing pKa prediction method from Simulations Plus based on artificial neural network ensembles (ANNE), microstates analysis, and literature data was retrained with a large homogeneous data set of drug-like molecules from Bayer. The new model was thus built with curated sets of ∼14,000 literature pKa values (∼11,000 compounds, representing literature chemical space) and ∼19,500 pKa values experimentally determined at Bayer Pharma (∼16,000 compounds, representing industry chemical space). Model validation was performed with several test sets consisting of a total of ∼31,000 new pKa values measured at Bayer. For the largest and most difficult test set with >16,000 pKa values that were not used for training, the original model achieved a mean absolute error (MAE) of 0.72, root-mean-square error (RMSE) of 0.94, and squared correlation coefficient (R(2)) of 0.87. The new model achieves significantly improved prediction statistics, with MAE = 0.50, RMSE = 0.67, and R(2) = 0.93. It is commercially available as part of the Simulations Plus ADMET Predictor release 7.0. Good predictions are only of value when delivered effectively to those who can use them. The new pKa prediction model has been integrated into Pipeline Pilot and the PharmacophorInformatics (PIx) platform used by scientists at Bayer Pharma. Different output formats allow customized application by medicinal chemists, physical chemists, and computational chemists.

Similarity-Based Classifier Using Topomers to Provide a Knowledge Base for hERG Channel Inhibition

In a proof of principle study we show that the similarity property principle can be applied to predict ADMET properties, exemplified on the case of hERG K+ channel inhibition. Early prediction of a drug candidates hERG channel activity is becoming increasingly important in the drug discovery process because blockade of the hERG channel may lead to life-threatening cardiac arrhythmias. Using the Tripos Topomer Search technology as compound similarity measure, we query molecules with unknown hERG activity for similar molecules in a set of compounds with known hERG activity. The hERG activity of the query molecule is then predicted based on the hERG activity of its Topomer Search neighbors and their distances to the query molecule. The similarity property principle can be applied with promising performance to predict hERG inhibition as long as there is a high structural overlap between the chemical spaces of the query compounds and the reference data set. We show that this is achievable for database sizes of about 10,000 structurally diverse molecules. In this case topoHERG is a similarity-based hERG classifier, which also acts as a knowledge base for hERG channel inhibition.

Utility of protein structures in overcoming ADMET-related issues of drug-like compounds

The number of solved X-ray structures of proteins relevant for ADMET processes of drug molecules has increased remarkably over recent years. In principle, this development offers the possibility to complement the quantitative structure-property relationship (QSPR)-dominated repertoire of in silico ADMET methods with protein-structure-based approaches. However, the complex nature and the weak nonspecific ligand-binding properties of ADMET proteins take structural biology methods and current docking programs to the limit. In this review we discuss the utility of protein-structure-based design and docking approaches aimed at overcoming issues related to plasma protein binding, active transport via P-glycoprotein, hERG channel mediated cardiotoxicity and cytochrome P450 inhibition, metabolism and induction.

SAM1 semiempirical calculations on the mechanism of cytochrome P450 metabolism

Abstract We present modeling results on the metabolism of alkanes by cytochrome P450 using the semiempirical method SAM1. For our model system consisting of unsubstituted porphyrin, iron, and methylthiolate, the structures for the complexes in all possible spin-states throughout the reaction cycle were calculated and extensive conformational searches were performed. These enable us to discuss the changes in relative energies, atomic charges and structures for each reaction step. The reaction cascade is exothermic for all steps except the last, the substrate oxidation, which has been suggested to be rate-limiting from experimental evidence. The oxygen–oxygen bond is elongated step-by-step and the atomic charge on the iron-bound oxygen analogously reduced. The porphyrin nitrogens act as a charge buffer, whereas iron becomes progressively more positive. Dioxygen is transformed to the singlet by end-on coordination to iron(II). The addition of two protons gives a concerted dissociation of water instead of the unfavorable water-oxide-complex. Thus, our results are in full agreement with experimental data and support the accepted mode of action of P450 enzymes.

Volume to dissolve applied dose (VDAD) and apparent dissolution rate (ADR): Tools to predict in vivo bioavailability from orally applied drug suspensions

Low solubility of drug candidates generated in research contributes to their elimination during subsequent development due to insufficient oral bioavailability (BA) of crystalline compound. Therefore, the purpose of the study was to identify critical in vitro solubility and dissolution parameter that would predict critical in vivo dissolution by means of in vitro-in vivo correlation. Thermodynamic solubility and apparent dissolution rate (ADR) were determined using the shake-flask method and mini-flow-through-cell, respectively. Oral BA studies in rats and humans were conducted from drug solution and suspension/tablets. Relative BA was calculated using F(rel) [%]=AUC(suspension)/AUC(solution)*100, representing a measure of in vivo dissolution. Roughly, F(rel) rat >50% translates into F(rel) human of >90%. Both, ADR and log volume to dissolve applied dose (VDAD), when plotted against F(rel) rat, revealed certain threshold levels, (ADR, ∼150-200 μg of compound dissolved under respective assay conditions; VDAD, ∼100-500 ml/kg) which translate into F(rel) in rats of >50%. Thus, assuming that F(rel)>50% in rats is indicative of sufficient in vivo dissolution in humans after oral application, drugs should exhibit a VDAD of ∼100-500 ml/kg or less in aqueous media to avoid insufficient or varying drug absorption.

Triphenylphosphonioacetylide: A Species Isoelectronic with Isocyanides

Stabilized only by triphenylphosphane, the title compound 1 was synthesized and investigated theoretically (the HOMO of 1 is depicted on the right). Both calculations and experimental NMR spectra show that no Lewis structure can describe the geometry and charge distribution accurately. Compound 1 undergoes an equivalent of the Passerini reaction. Ph3 P+ -C≡C- : 1.

Site of Metabolism Prediction Based on abinitio Derived Atom Representations

Machine learning models for site of metabolism (SoM) prediction offer the ability to identify metabolic soft spots in low‐molecular‐weight drug molecules at low computational cost and enable data‐based reactivity prediction. SoM prediction is an atom classification problem. Successful construction of machine learning models requires atom representations that capture the reactivity‐determining features of a potential reaction site. We have developed a descriptor scheme that characterizes an atoms steric and electronic environment and its relative location in the molecular structure. The partial charge distributions were obtained from fast quantum mechanical calculations. We successfully trained machine learning classifiers on curated cytochrome P450 metabolism data. The models based on the new atom descriptors showed sustained accuracy for retrospective analyses of metabolism optimization campaigns and lead optimization projects from Bayer Pharmaceuticals. The results obtained demonstrate the practicality of quantum‐chemistry‐supported machine learning models for hit‐to‐lead optimization.

Femtosecond dynamics in the lactim tautomer of phycocyanobilin: a long-wavelength absorbing model compound for the phytochrome chromophore.

Transient UV/Vis absorption spectroscopy is used to study the primary dynamics of the ring-A methyl imino ether of phycocyanobilin (PCB-AIE), which was shown to mimic the far-red absorbance of the Pfr chromophore in phytochromes (R. Micura, K. Grubmayr, Bioorg. Med. Chem. Lett.- 1994, 4, 2517-2522). After excitation at 615 nm, the excited electronic state is found to decay with τ1 =0.4 ps followed by electronic ground-state relaxation with τ2 =1.2 and τ3 =6.7 ps. Compared with phycocyanobilin (PCB), the initial kinetics of PCB-AIE is much faster. Thus, the lactim structure of PCB-AIE seems to be a suitable model that could not only explain the bathochromic shift in the ground-state absorption but also the short reaction of the Pfr as compared to the Pr chromophore in phytochrome. In addition, the equivalence of ring-A and ring-D lactim tautomers with respect to a red-shifted absorbance relative to the lactam tautomers is demonstrated by semiempirical calculations.

1H-Phosphiranes and -enes invert at phosphorusvia a turnstile rotation

Ab initio (CCSD(T)/6-31+G*) calculations suggest that 1H-phosphiranes and 1H-phosphirenes invert their configuration at phosphorus by a rotation of the PX group (X = H, F, Cl, Br) above the C 2 moiety, rather than the more usual planar trigonal inversion pathway via a C 2v transition state.