Virtual Molecules

Proteolysis targeting chimeras represent a class of drug molecules with a number of attractive properties, most notably a potential to work for tar- gets that, so far, have been in-accessible for conventional small molecule inhibitors. Due to their different mechanism of action, and physico-chemical properties, many of the methods that have been designed and applied for computer aided design of traditional small molecule drugs are not applica- ble for proteolysis targeting chimeras. Here we review recent developments in this field focusing on three aspects: de-novo linker-design, estimation of absorption for beyond-rule-of-5 compounds, and the generation and ranking of ternary complex structures. In spite of this field still being young, we find that a good number of models and algorithms are available, with the poten- tial to assist the design of such compounds in-silico, and accelerate applied pharmaceutical research.

Proximity-inducing compounds (PICs) are an emergent drug technology through which a protein of interest (POI), often a drug target, is brought into the vicinity of a second protein which modifies the POI's function, abundance or localisation, giving rise to a therapeutic effect. One of the best-known examples for such compounds are heterobifunctional molecules known as proteolysis targeting chimeras (PROTACs). PROTACs reduce the abundance of the target protein by establishing proximity to an E3 ligase which labels the protein for degradation via the ubiquitin-proteasomal pathway. Design of PROTACs in silico requires the computational prediction of the ternary complex consisting of POI, PROTAC molecule, and the E3 ligase. We present a novel machine learning-based method for predicting PROTAC-mediated ternary complex structures using Bayesian optimization. We show how a fitness score combining an estimation of protein-protein interactions with PROTAC conformation energy calculations enables the sample-efficient exploration of candidate structures. Furthermore, our method presents two novel scores for filtering and reranking which take PROTAC stability (Autodock-Vina based PROTAC stability score) and protein interaction restraints (the TCP-AIR score) into account. We evaluate our method using DockQ scores on a number of available ternary complex structures (including previously unevaluated cases) and demonstrate that even with a clustering that requires members to have a high similarity, i.e., with smaller clusters, we can assign high ranks to those clusters that contain poses close to the experimentally determined native structure of the ternary complexes. We also demonstrate the resultant improved yield of near-native poses in these clusters.

L-Lactate oxidase (LOX) belongs to a large family of flavoenzymes that catalyze oxidation of α-hydroxy acids. How in these enzymes the protein structure controls reactivity presents an important but elusive problem. LOX contains a prominent tyrosine in the substrate binding pocket (Tyr(215) in Aerococcus viridans LOX) that is partially responsible for securing a flexible loop which sequesters the active site. To characterize the role of Tyr(215), effects of substitutions of the tyrosine (Y215F, Y215H) were analyzed kinetically, crystallographically and by molecular dynamics simulations. Enzyme variants showed slowed flavin reduction and oxidation by up to 33-fold. Pyruvate release was also decelerated and in Y215F, it was the slowest step overall. A 2.6-Å crystal structure of Y215F in complex with pyruvate shows the hydrogen bond between the phenolic hydroxyl and the keto oxygen in pyruvate is replaced with a potentially stronger hydrophobic interaction between the phenylalanine and the methyl group of pyruvate. Residues 200 through 215 or 216 appear to be disordered in two of the eight monomers in the asymmetric unit suggesting that they function as a lid controlling substrate entry and product exit from the active site. Substitutions of Tyr(215) can thus lead to a kinetic bottleneck in product release.

Histone deacetylase 3 (HDAC3) is a promising epigenetic drug target for multiple therapeutic applications. Direct interaction between the Deacetylase Activating Domain of the silencing mediator for retinoid or thyroid-hormone receptors (SMRT-DAD) is required for activation of enzymatic activity of HDAC3. The structure of this complex and the nature of interactions with HDAC inhibitors in solution are unknown. Using novel photoreactive HDAC probes, “nanorulers”, we determined the distance between the catalytic site of the full-length HDAC3 and SMRT-DAD in solution at physiologically relevant conditions and found it to be substantially different from that predicted by the X-ray model with a Δ379–428 aa truncated HDAC3. Further experiments indicated that in solution this distance might change in response to chemical stimuli, while the enzymatic activity remained unaffected. These observations were further validated by Saturation Transfer Difference (STD) NMR experiments. We propose that the observed changes in the distance are an important part of the histone code that remains to be explored. Mapping direct interactions and distances between macromolecules with such “nanorulers” as a function of cellular events facilitates better understanding of basic biology and ways for its manipulation in a cell- and tissue-specific manner.

Hyperactivation of the calcium-dependent cysteine protease calpain 1 (Cal1) is implicated as a primary or secondary pathological event in a wide range of illnesses and in neurodegenerative states, including Alzheimer’s disease (AD). E-64 is an epoxide-containing natural product identified as a potent nonselective, calpain inhibitor, with demonstrated efficacy in animal models of AD. By use of E-64 as a lead, three successive generations of calpain inhibitors were developed using computationally assisted design to increase selectivity for Cal1. First generation analogues were potent inhibitors, effecting covalent modification of recombinant Cal1 catalytic domain (Cal1cat), demonstrated using LC–MS/MS. Refinement yielded second generation inhibitors with improved selectivity. Further library expansion and ligand refinement gave three Cal1 inhibitors, one of which was designed as an activity-based protein profiling probe. These were determined to be irreversible and selective inhibitors by kinetics studies comparing full length Cal1 with the general cysteine protease papain.

The active site of mannitol 2-dehydrogenase from Pseudomonas fluorescens (PfM2DH) is connected with bulk solvent through a narrow protein channel that shows structural resemblance to proton channels utilized by redox-driven proton pumps. A key element of the PfM2DH channel is the “mobile” Glu292, which was seen crystallog. to adopt distinct positions up and down the channel. It was suggested that the “down up” conformational change of Glu292 could play a proton relay function in enzymic catalysis, through direct proton shuttling by the Glu or because the channel is opened for water mols. forming a chain along which the protons flow. We report evidence from site-directed mutagenesis (Glu292 Ala) substantiated by data from mol. dynamics simulations that support a role for Glu292 as a gate in a water chain (von Grotthuss-type) mechanism of proton translocation. Occupancy of the up and down position of Glu292 is influenced by the bonding and charge state of the catalytic acid base Lys295, suggesting that channel opening/closing motions of the Glu are synchronized to the reaction progress. Removal of gatekeeper control in the E292A mutant resulted in a selective, up to 120-fold slowing down of microscopic steps immediately preceding catalytic oxidn. of mannitol, consistent with the notion that formation of the productive enzyme-NAD+-mannitol complex is promoted by a corresponding position change of Glu292, which at physiol. pH is assocd. with obligatory deprotonation of Lys295 to solvent. These results underscore the important role of conformational dynamics in the proton transfer steps of alc. dehydrogenase catalysis.

UDP-xylose synthase (UXS) catalyzes decarboxylation of UDP-D-glucuronic acid to UDP-xylose. In mammals, UDP-xylose serves to initiate glycosaminoglycan synthesis on the protein core of extracellular matrix proteoglycans. Lack of UXS activity leads to a defective extracellular matrix, resulting in strong interference with cell signaling pathways. We present comprehensive structural and mechanistic characterization of the human form of UXS. The 1.26 ?? crystal structure of the enzyme bound with NAD(+) and UDP reveals a homodimeric short-chain dehydrogenase/reductase (SDR), belonging to NDP-sugar epimerases/dehydratases subclass. We show that enzymatic reaction proceeds in three chemical steps via UDP-4-keto-D-glucuronic acid and UDP-4-keto-pentose intermediates. Molecular dynamics simulations reveal that the D-glucuronyl ring accommodated by UXS features a marked (4)C(1) chair to B(O,3) boat distortion that facilitates catalysis in two different ways. It promotes oxidation at C4 (step 1) by aligning the enzymatic base Tyr(147) with the reactive substrate hydroxyl and it brings the carboxylate group at C(5) into an almost fully axial position, ideal for decarboxylation of UDP-4-keto-D-glucuronic acid in the second chemical step. The protonated side chain of Tyr(147) stabilizes the enolate of decarboxylated C(4) keto species ((2)H(1 ) half-chair) that is then protonated from re face at C(5), involving water coordinated by Glu(120). Arg(277), which is positioned by a salt-link interaction with Glu(120), closes up the catalytic site and prevents release of the UDP-4-keto-pentose and NADH intermediates. Hydrogenation of the C(4) keto group by NADH, assisted by Tyr(147) as catalytic proton donor, yields UDP-xylose adopting the relaxed (4)C(1) chair conformation (step 3).

A systematic investigation of the available crystal structures of HDAC8 and of the influence of different receptor structures and docking protocols is presented. The study shows that the open conformation of HDAC8 may be preferred by ligands with flexible surface binding groups, as such a conformation allows the ligands to minimize their exposure to solvent upon binding. This observation allowed us to rationalize the excellent potency of pyrazole-based inhibitors compared to that of isoxazole-based inhibitors.

The design, synthesis, docking, and biol. evaluation of novel potent HDAC3 and HDAC8 isoxazole- and pyrazole-based diazide probes suitable for binding ensemble profiling with photoaffinity labeling (BEProFL) expts. in cells is described. Both the isoxazole- and pyrazole-based probes exhibit low nanomolar inhibitory activity against HDAC3 and HDAC8, resp. The pyrazole-based probe I appears to be one of the most active HDAC8 inhibitors reported in the literature with an IC50 of 17 nM. Our docking studies suggest that unlike the isoxazole-based ligands the pyrazole-based ligands are flexible enough to occupy the second binding site of HDAC8. Probes/inhibitors II, III (R = H, N3) and I exerted the antiproliferative and neuroprotective activities at micromolar concns. through inhibition of nuclear HDACs, indicating that they are cell permeable and the presence of an azide or a diazide group does not interfere with the neuroprotection properties, or enhance cellular cytotoxicity, or affect cell permeability.

A binding ensemble profiling with (f)photoaffinity labeling (BEProFL) approach that utilizes photolabeling of HDAC8 with a probe contg. a UV-activated arom. azide, mapping of the covalent modifications by liq. chromatog.-tandem mass spectrometry, and a computational method to characterize the multiple binding poses of the probe is described. By use of the BEProFL approach, two distinct binding poses of the HDAC8 probe were identified. The data also suggest that an “upside-down” pose with the surface binding group of the probe bound in an alternative pocket near the catalytic site may contribute to the binding.

Pantothenate synthetase (PS) is one of the potential new antimicrobial targets that may also be useful for the treatment of the nonreplicating persistent forms of Mycobacterium tuberculosis. In this Letter we present a series of 5-tert-butyl-N-pyrazol-4-yl-4,5,6,7-tetrahydrobenzo[d]isoxazole-3-carboxamide derivs. (e.g. I) as novel potent Mycobacterium tuberculosis PS inhibitors, their in silico mol. design, synthesis, and inhibitory activity.

The mechanical properties of crystalline materials can have an impact on different aspects of solid dosage form design and manufacturing. Experimental determination of these quantities is known to pose difficulties and can be time-consuming. Instead, first-principles predictions of such materials have the potential for accelerating drug formulation development. We calculated elastic constants and several derived mechanical properties for the crystalline phase of a set of 18 small organic molecules with available experimental data, using molecular dynamics simulations based on the widely used GAFF2 force field. Attempts to further improve the accuracy of the predicted mechanical properties were made by fitting the force field parameters to the energies and forces determined by DFT calculations. We determine that DFT-based optimization of the force field parameters were unable to improve the accuracy of the mechanical property predictions over the currently established GAFF2 force field, which already provides reasonably accurate results with mean unsigned errors of individual elastic constants of around 2-3 GPa, or lower, in the majority of cases. Also, the elastic anisotropy is correctly predicted, at least qualitatively. The method used here can be a simple and fast means for the approximate determination of mechanical properties of a crystalline drug powder-material properties that can be used to guide decisions during formulation development.

Oxidative degradation of drugs is one of the major routes of drug substance and drug product instability. Among the diverse routes of oxidation, autoxidation is considered to be challenging to predict and control, potentially due to the multi-step mechanism involving free radicals. C-H bond dissociation energy (C-H BDE) is evidenced to be a calculated descriptor shown to predict drug autoxidation. While computational predictions for the autoxidation propensity of drugs are both swift and possible, no literature to date has highlighted the relationship between the computed C-h BDE and the experimentally-derived autoxidation propensities of solid drugs. The objective of this study is to investigate this missing relationship. The present work is an extension to the previously reported novel autoxidation approach that involves subjecting a physical mixture of pre-milled polyvinyl pyrrolidone (PVP) K-60 and a crystalline drug under high temperature and pressurized oxygen setup. The drug degradation was measured using chromatographic methods. An improved trend between the extent of solid autoxidation and C-H BDE could be observed after normalizing the effective surface area of drugs in the crystalline state, pointing to a positive relationship. Additional studies were conducted by dissolving the drug in N-methyl pyrrolidone (NMP) and exposing the solution under a pressurized oxygen setup at diverse elevated temperatures. Chromatographic results of these samples indicated a similarity in the formed degradation products to the solid-state experiments pointing to the utility of NMP, a PVP monomer surrogate, as a stressing agent for faster and relevant autoxidation screening of drugs in formulations.

Common energy-intensive processes applied in oral solid dosage development, such as milling, sieving, blending, compaction, etc. generate particles with surface and bulk crystal disorder. An intriguing aspect of the generated crystal disorder is its evolution and repercussion on the physical- and chemical stabilities of drugs. In this review, we firstly examine the existing literature on crystal disorder and its implications on solid-state stability of pharmaceuticals. Secondly, we discuss the key aspects related to the generation and evolution of crystal disorder, dynamics of the disordered/amorphous phase, analytical techniques to measure/quantify them, and approaches to model the disordering propensity from first principles. The main objective of this compilation is to provide special impetus to predict or model the chemical degradation(s) resulting from processing-induced manifestation in bulk solid manufacturing. Finally, a generic workflow is proposed that can be useful to investigate the relevance of crystal disorder on the degradation of pharmaceuticals during stability studies. The present review will cater to the requirements for developing physically- and chemically stable drugs, thereby enabling early and rational decision-making during candidate screening and in assessing degradation risks associated with formulations and processing.

The C-H bond dissociation energy (BDE) of drug molecules is often used to estimate their relative propensities to undergo autoxidation. BDE calculations based on electronic structures provide a convenient means to estimate the risk for a given compound to degrade via autoxidation. This study aimed to verify the utility of calculated C-H BDEs of a range of drug molecules in predicting their autoxidation propensities, in the solution state. For the autoxidation study, 2,2-azobis (2-methylpropionitrile) was employed as the solution state stressor, and the experimental reaction rate constants were determined employing ultraperformance liquid chromatographic (UPLC) methods. Reaction rates in the solution state were compared to the calculated C-H BDE values of the respective compounds. The results indicated a poor correlation for compounds in the solution state, and their relative stabilities could not be explained with C-H BDE. On the other hand, a favorable relationship was observed between the relative extent of ionization and the autoxidation rates of the selected compounds. In the solution state, factors such as the type and extent of drug ionization, degree and type of solvation have been shown to contribute to differences in reactivity. By applying the computational method involving the effect of H-atom abstraction and potential ionization sites in the molecule, the calculated C-H BDE should relate better to the experimental autoxidation rates.

Co-amorphous systems (COAMS) have raised increasing interest in the pharmaceutical industry, since they combine the increased solubility and/or faster dissolution of amorphous forms with the stability of crystalline forms. However, the choice of the co-former is critical for the formation of a COAMS. While some models exist to predict the potential formation of COAMS, they often focus on a limited group of compounds. Here, four classes of combinations of an active pharmaceutical ingredient (API) with (1) another API, (2) an amino acid, (3) an organic acid, or (4) another substance were considered. A model using gradient boosting methods was developed to predict the successful formation of COAMS for all four classes. The model was tested on data not seen during training and predicted 15 out of 19 examples correctly. In addition, the model was used to screen for new COAMS in binary systems of two APIs for inhalation therapy, as diseases such as tuberculosis, asthma, and COPD usually require complex multidrug-therapy. Three of these new API-API combinations were selected for experimental testing and co-processed via milling. The experiments confirmed the predictions of the model in all three cases. This data-driven model will facilitate and expedite the screening phase for new binary COAMS

In the present study, the oxidative degradation behavior of nifedipine (NIF) in amorphous solid dispersions (ASDs) prepared with poly(vinyl pyrrolidone) (PVP) with a short (K30) and a long (K90) chain length was investigated. The ASDs were prepared via dry ball-milling and analyzed using Fourier transform infrared (IR) spectroscopy, X-ray scattering, and differential scanning calorimetry. The ASDs were exposed to accelerated thermal-oxidative conditions using a pressurized oxygen headspace (120 °C for 1 day) and high temperatures at atmospheric pressure (60–120 °C for a period of 42 days). Additionally, solution-state oxidative degradation studies showed that pure NIF degrades to a greater extent than in the presence of PVP. Electronic structure calculations were performed to understand the impact of drug–polymer intermolecular interactions on the autoxidation of drugs. While no drug degradation was observed in freshly prepared ASD samples, alkyl free radicals were detected via electron paramagnetic resonance (EPR) spectroscopy. The free radicals were found to be consumed to a greater extent by PVP K30- than PVP K90-based ASDs upon exposure to high oxygen pressures. This was consistent with the greater solid-state oxidative degradation of NIF observed in ASDs with PVP K30 than with PVP K90. As no drug recrystallization occurred during this study period, the lower glass-transition temperature and presumed greater molecular mobility of PVP K30 and its ASD as compared to the PVP K90 system appear to contribute to the greater drug degradation in PVP-K30-based ASDs. The extent and the rate of oxidative degradation were higher in the case of PVP-K30-based ASD as compared to that in PVP-K90-based ASD, and the overall degradation increased with an increase in temperature. IR spectral analysis of drug–polymer interactions supports the electronic calculations of the oxidation process. We infer that, apart from the initial free radical content, the difference in the extent of drug–polymer intermolecular interactions in ASDs and amorphous stabilization during the forced oxidation experiments contribute to the observed differences in the autoxidative reactivity of the drug in ASDs with different PVP chain lengths. Overall, the chemical degradation of NIF in ASDs with two PVP chain lengths obtained from accelerated solid-state oxidation studies was in qualitative agreement with that obtained from long-term (3 years) storage under ambient conditions. The study highlights the ability of accelerated processes to determine the oxidative degradation behavior of polymeric ASDs and suggests that the polymer chain length could factor into chemical as well as physical stability considerations.

The present study aimed to investigate methods for accelerating autoxidation of crystalline drugs in the solid-state that can potentially predict real-time stability. Solid droperidol (DPD) was selected as the model drug. A common free-radical initiator, 2,2'-azobisisobutyronitrile (AIBN), was used to induce autoxidation in solutions. AIBN decomposes at elevated temperatures to yield carbon-centred cyano-isopropyl free radicals that can auto-oxidize neighboring drug molecules. Although the reaction of AIBN is relatively straightforward in solution, it is less so in solids. In this study, we used solid AIBN mixed with DPD powder in the presence and absence of pressurized oxygen headspace. Samples were prepared directly in the form of binary mixtures with DPD and additionally in the form of powder compact/pellet with DPD. The main challenge in carrying out the reaction was related to the preservation of AIBN at elevated temperatures due to the disintegration of the pellet containing the latter. A commercially available free-radical coated silica particle (i.e., 2,2,6,6-tetramethyl-1-piperinyloxy (TEMPO) or (SiliaCATTM TEMPO)) was tested as a potential stressor, but with limited success to induce autoxidation. The most valuable results were obtained when a physical mixture of pre-milled PVP K-60 containing free radicals and DPD was exposed to elevated oxygen-temperature conditions, which yielded significant degradation of DPD. The study highlights the practical challenges for conducting accelerated solid-state stress studies to assess the autoxidation susceptibility of drugs using traditional free-radical initiators and presents a proof of application of milled PVP with free-radical as a potential alternative.

The present study investigated the impact of solid-state disorders generated during milling on the chemical reactivity of simvastatin. An amorphous and a partially crystalline simvastatin powders were generated via cryomilling simvastatin crystals for either 90 or 10 min, respectively. The thoroughly characterized milled powders were stored at 40°C/75% RH, in open and closed containers. The effect of milling and storage conditions on physical stability was investigated using simultaneous small and wide-angle X-ray scattering and differential scanning calorimetry. The chemical degradation was evaluated using liquid chromatography-mass spectrometry. Compared with the fully amorphous state, the partially crystalline simvastatin crystallized to a lower extent in the expense of higher chemical degradation on open storage. The closely stored samples degraded to a lower extent and crystallized to a higher extent than the openly stored ones. However, the trends of the total crystallinity and degradation between amorphous and partially crystalline powders were similar. Small-angle X-ray scattering revealed that the partially crystalline simvastatin comprised a higher extent of nanoscale density heterogeneity than the fully amorphous powder. The overall results pointed toward the role of the remaining amorphous content and the nanoscale and mesoscale density heterogeneity on the chemical reactivity in the disordered simvastatin.

Purpose: The performance of carrier-based dry powder inhaler (DPI) formulations can be critically impacted by interfacial interactions driven by tribo-electrification. Therefore, the aim of the present work was to understand how distinct API particle characteristics affect the charging behaviour of blends intended for DPI delivery. Methods: Salbutamol sulphate (SBS) particles engineered via spray-drying and jet milling were used as model APIs. D-mannitol was selected as a model carrier. The materials were characterized concerning their different particle properties and their charge was analysed alone and in blends before and after flow over a stainless-steel pipe. Results: The spray-dried SBS (amorphous and spherical) charged positively and to a higher extent than jet milled SBS (crystalline and acicular) that charged negatively and to a lower extent. D-mannitol charged positively and to a higher extent than the APIs. All drug-excipient blends charged negatively and differences were found between the spray-dried and jet milled SBS blends at 2% and 5% drug loads. Conclusions: It was demonstrated how distinct solid-states, particle shape, size and morphology as well as different water contents of the different materials can affect tribo-charging. For their binary blends, the amount and nature of fines seem to govern inter-particle contacts critically impacting charge evolution.

Log Po​/w based models are often used for the retention time prediction of reversed phase liq. chromatog. Here, we present the investigation of the applicability of log Po​/w based retention time predictions for the sepn. in capillary electro-​chromatog. (CEC)​. A test set of five polycyclic arom. hydrocarbons was sepd. using two different stationary phases with three different mobile phases each. The resulting retention times were correlated with the exptl. log Po​/w values as well as with calcd. log Po​/w values. The used methods include quant. structure property relationship (QSPR) models as well as mol. dynamic methods such as the linear interaction energy (LIE) or the Bennett acceptance ratio (BAR)​. The results indicate that rigorous simulation models are capable of accurately reproducing exptl. results and that the electrophoretic mobility of analytes in CEC sepns. leads to significant deviations in the retention time prediction.

The interaction between drug products and polymeric packaging materials is an important topic in the pharmaceutical industry and often associated with high costs because of the required elaborative interaction studies. Therefore, a theoretical prediction of such interactions would be beneficial. Often, material parameters such as the octanol water partition coefficient are used to predict the partitioning of migrant molecules between a solvent and a polymeric packaging material. Here, we present the investigation of the partitioning of various migrant molecules between polymers and solvents using molecular dynamics simulations for the calculation of interaction energies. Our results show that the use of a model for the interaction between the migrant and the polymer at atomistic detail can yield significantly better results when predicting the polymer solvent partitioning than a model based on the octanol water partition coefficient.

The interaction between packaging materials and drug products is an important issue for the pharmaceutical industry, since during manufg., processing and storage a drug product is continuously exposed to various packaging materials. The exptl. investigation of a great variety of different packaging material-drug product combinations in terms of efficacy and safety can be a costly and time-consuming task. In our work we used mol. dynamics (MD) simulations in order to evaluate the applicability of such methods to pre-screening of the packaging material-solute compatibility. The solvation free energy and the free energy of adsorption of diverse solute/solvent/solid systems were estd. The results of our simulations agree with exptl. values previously published in the literature, which indicates that the methods in question can be used to semi-quant. reproduce the solid-liq. interactions of the investigated systems.

Mechanical properties of crystalline materials can have an impact on different aspects of solid dosage form design and manufacturing. Experimentally determination of these quantities is known to pose difficulties and can be time consuming. Instead, first principles predictions of such materials have the potential for accelerating drug formulation development. We calculated elastic constants and several derived mechanical properties for the crystalline phase of a set of 18 small organic molecules with available experimental data, using Molecular Dynamics simulations based on the widely used GAFF2 force field. Attempts to further improve the accuracy of the predicted mechanical properties were made by fitting the force field parameters to the energies and forces determined by DFT calculations. We determine that DFT based optimization of the force field parameters were unable to improve the accuracy of the mechanical property predictions over the currently established GAFF2 force field, which already provides reasonably accurate results with mean unsigned errors of individual elastic constants around 2-3 GPa, or lower, in the majority of cases. Also, the elastic anisotropy is correctly predicted, at least qualitatively. The method used here can be a simple and fast means for the approximate determination of mechanical properties of a crystalline drug powder - material properties that can be used to guide decisions during formulation development.

Tribo-charging can negatively impact safety and efficacy of pharmaceutical products due to particle agglomeration and adherence to surfaces resulting in poor mixing and inaccurate dosing. Prediction of relative charging trends in-silico and from first principles could provide an efficient tool for guiding development during early formulation studies. We experimentally investigate tribo-charging of pharmaceutical compounds and compare the results with energies from electronic-structure-calculations. Calculations were performed applying different levels of theory (semi-empirical and DFT with different functionals) to estimate properties including ionization potentials (IP), and work functions (WF). The calculations are performed for single molecules assuming that, at least, relative properties of the solid phase are reproduced. Our results suggest that, independently of the level of theory, calculated IP values correlate well with experimental tribo-charging trends provided we use the adiabatic IP. By comparison, WF values, calculated by a commonly used approximation, turn out being a poor descriptor for tribo-charging trends.

Amorphous solid dispersions are considered a promising formulation strategy for the oral delivery of poorly soluble drugs. The limiting factor for the applicability of this approach is the physical (in)stability of the amorphous phase in solid samples. Minimizing the risk of reduced shelf life for a new drug by establishing a suitable excipient/polymer-type from first principles would be desirable to accelerate formulation development. Here, we perform Molecular Dynamics simulations to determine properties of blends of eight different polymer–small molecule drug combinations for which stability data are available from a consistent set of literature data. We calculate thermodynamic factors (mixing energies) as well as mobilities (diffusion rates and roto-vibrational fluctuations). We find that either of the two factors, mobility and energetics, can determine the relative stability of the amorphous form for a given drug. Which factor is rate limiting depends on physico-chemical properties of the drug and the excipients/polymers. The methods outlined here can be readily employed for an in silico pre-screening of different excipients for a given drug to establish a qualitative ranking of the expected relative stabilities, thereby accelerating and streamlining formulation development

Understanding properties and mechanisms that govern drug degradation in the solid state is of high importance to ensure drug stability and safety of solid dosage forms. In this study, we attempt to understand drug-excipient interactions in the solid state using both theoretical and experimental approaches. The model active pharmaceutical ingredients (APIs) under study are carvedilol (CAR) and codeine phosphate (COP), which are known to undergo esterification with citric acid (CA) in the solid state. Starting from the crystal structures of two different polymorphs of each compound, we calculated the exposure and accessibility of reactive hydroxyl groups for a number of relevant crystal surfaces, as well as descriptors that could be associated with surface stabilities using molecular simulations. Accelerated degradation experiments at elevated temperature and controlled humidity were conducted to assess the propensity of different solid forms of the model APIs to undergo chemical reactions with anhydrous CA or CA monohydrate. In addition, for CAR, we studied the solid state degradation at varying humidity levels and also under mechano-activation. Regarding the relative degradation propensities, we found that variations in the exposure and accessibility of molecules on the crystal surface play a minor role compared to the impact of molecular mobility due to different levels of moisture. We further studied drug-excipient interactions under mechano-activation (comilling of API and CA) and found that the reaction proceeded even faster than in physical powder mixtures kept at accelerated storage conditions.

The aggregation of Proteins is a phenomenon of considerable interest, both in fundamental and applied pharmaceutical research. Next to its role in the field of neuro degenerative diseases it comprises one of the major challenges for the successful formulation of protein based parental drugs. The second virial coefficient (B22) has been shown to correlate with the stability of a protein solutions in a number of cases. Here we use a simple pseudo protein model to rigorously test different simulation strategies for the calculatuion of B22 values, comparing their accuracy and computational efficiency. We also present results obtained with more realistic protein models and discuss the predictive power and the limitations of such models.

A predictive model of the effect of crystal agglomeration on particle form and size distribution requires the quantification of various process parameters that depend on the microscopic properties of specific crystal faces and their interaction with the solvent. The authors discuss the various stages in the agglomeration process, using the results of recent expts. on breaking the agglomerative bond and at. level simulations on the forces involved in crystal aggregation, to highlight the questions that need to be resolved for agglomeration processes to be understood.

A novel exptl. app. has been developed which enables the measurement of adhesion forces between two crystals suspended in a supersatd. soln. and allowed to agglomerate over a fixed time period. The geometry of the crystal surfaces at the contact points and the dynamic development of the bond are captured on video and characterized using an image anal. technique. The exptl. app. has been designed to allow control of supersatn., orientation of crystal faces, distance between crystals, relative movement of crystals and contact time. The exptl. results show that the agglomerate bond strength, expressed as the agglomerate adhesion force per unit contact area, increases with increasing supersatn. and is higher for faster growing faces than for slower growing faces. In addn., a qual. comparison has been made between the measured force and a theor. estn. of the interaction force between crystal faces, detd. through mol. modeling. It is shown that the speed of approach of two opposing crystal faces is a key parameter in the nature of the subsequent bond, as is their at. structure.

The theor. investigation of polar crystal surfaces in soln. involves methodol. issues that have not been entirely resolved. Here we present results of mol. dynamics simulations of potash alum surfaces in contact with aq. soln. As a prerequisite for this study, we developed a classical, empirical model potential for this complex inorg. material. The resulting rigid ion force field was carefully tested, and we could show its ability to reproduce a wide range of exptl. results. Using this force field, we studied the (111) surface of potash alum and compared various ways of simulating this polar surface. By establishing the stability of a charged, in addn. to the reconstructed crystal surface usually assumed, we question the common practice of excluding the existence of such charged surface terminations a priori.

The influence of the intermol. potential on predicted growth morphologies is studied. The authors compare the morphologies calcd. with the attachment energy model for Î?-caprolactam (C2/c), Î?-succinic acid (P21/c), hexamethylenetetramine (Iâ??43m), pentaerythritol (Iâ??4), and urea (Pâ??421m). Five different model potentials that were developed for modeling org. crystal structures are used. These show a progressive improvement in the accuracy of the electrostatic model and a variety of empirical parametrizations of the repulsion-dispersion terms. Although the lattice energies and abs. attachment energies vary quite significantly, there are only minor variations in the predicted morphologies. The assumptions in the attachment energy model generally contribute more to any discrepancy between the predicted and obsd. morphologies than the variations between different model potentials that could be considered appropriate for modeling. However, the attachment energy model readily provides a reasonable est. of the dominant faces of the vapor-grown habit, suitable for most chem. engineering purposes.

Use of atom based truncation schemes in thermodn. integration leads to hydration free energies of water that depend strongly on the cut-off radius. By contrast, Ewald summation, a reaction field technique, and group based truncation of the potential give the same result (6.5Â?0.1 kcal/mol), independently of the cut-off radius and system size.

The control of protein aggregation is an important requirement in the development of bio-pharmaceutical formulations. Here a simple protein model is proposed that was used in molecular dynamics simulations to obtain a quantitative assessment of the relative contributions of proteins’ net-charges, dipole-moments, and the size of hydrophobic or charged surface patches to their colloidal interactions. The results demonstrate that the strength of these interactions correlate with net-charge and dipole moment. Variation of both these descriptors within ranges typical for globular proteins have a comparable effect. By comparison no clear trends can be observed upon varying the size of hydrophobic or charged patches while keeping the other parameters constant. The results are discussed in the context of experimental literature data on protein aggregation. They provide a clear guide line for the development of improved algorithms for the prediction of aggregation propensities.

The control of protein aggregation is an important requirement in the development of bio-pharmaceutical formulations. Here a protein model is proposed that combines a rigid spherical geometry with a heterogeneous surface charge distribution based on such distributions observed on small globular proteins. The model is used in Molecular Dynamics simulations in explicit solvent to establish the relative contributions of net-charge, dipole-moment and hydrophobicity to colloidal interactions. The results suggest that the strength of these interactions correlate with, both, net-charge and dipole moment. By comparison no clear trends can be observed when varying the size of hydrophobic patches while keeping the other parameters constant. The results are discussed in the context of experimental literature data on protein aggregation, and implications for the development of algorithms for the prediction of aggregation propensities are proposed.

A broad consensus exists in the TB research community that the key to shortening the duration of drug therapy, and to decreasing relapse rates, lies in the specific targeting of the pathways essential to the non-replicating persistence (NRP) state of Mtb. One such target is malate synthase (MS or GlcB), an enzyme in glyoxylate shunt. It has been shown that disruption of the glyoxylate shunt attenuates bacterial persistent and virulence in immunocompetent mice. Recent success in the structure elucidation of the MS protein and the availability of functional assays for MS opens a way to apply computer-aided drug design methods in concert with medicinal chem. efforts to discover new therapeutic agents for NRP TB. The structure-based and ligand-based drug design methods were used to generate leads with chem. diversity. 95 compds. were tested in vivo and in vitro against malate synthase. At 60 uM concn., one compd. inhibited more than 90% of MS activity, and two compds. inhibited around 60%. At 300 uM concn., one compd. inhibited more than 70% of MS activity, one compd. inhibited around 40%, eight compds. more than 20%, and four compds. more than 10% of activity of MS. In MABA assay, 8 compds. having % Inhibition > 70% at 128 uM and 12 compds. having % Inhibition > 70% at Conc. 100 ug/mL. 9 compds. have a MIC value less than 100 ug/mL in LORA. 10 compds. that were active in both MABA and LORA were tested against Vero cell toxicity with 5 compds. having an IC50s ranging from ca 5-70 uM. In this study the design, SAR, computational anal., and future plans will be discussed.

abstract

The majority of the modern docking software is able to reproduce the native position of the ligands in the active site. However, the accuracy of the docking measured by RMSD between the native and docked ligands declines if the binding sites of interest are flexible or solvent exposed. In this study we assessed the accuracy of GOLD and Surflex in combination with MMPBSA in re-docking of the ligands in 55 NMR protein-ligand complexes. The accuracy of the docking poses is discussed w.r.t min. RMSD, cRMSD of atoms in active site, and atoms-pairs-based interaction fingerprint (APIF). The results of our study suggest that the usage and assessment of the docking and scoring functions for the solvent exposed ligands may need different criteria compared to what is commonly used in the out-of-the-box docking software.

In this presentation, we describe the design, synthesis, and biol. evaluation of active and selective HDAC3 and HDAC8 diazide isoxazole-based and pyrazole-based probes suitable for photolabeling expts. in cells. We also performed a closely related small SAR study focused on a novel pyrazole-based scaffold and optimized the synthetic procedures for gram scale synthesis. Both the isoxazole- and pyrazole-based probes exhibit excellent low double digit nanomolar inhibitory activity against HDAC3 and HDAC8, resp. The cell-based expts. show that these probes are cell permeable and exert an anti-proliferative activity towards HepG2, Hela, and SH-SY5Y cell lines, and neuroprotective activity in SH-SY5Y cell lines at micromolar concns. The presence of an azide or a diazide group does not interfere with the neuroprotection properties, or enhance cellular cytotoxicity, or affect cell permeability, making these probes excellent candidates for cell-based photolabeling expts

Protein binding sites can include conserved water mols. that mediate, in some cases crucial, interactions between receptor and ligands. The no. and identity of such water mols. may vary depending on the ligand considered. Here we try to account for this fact by performing docking calcns. with multiple binding sites, differing in their water occupancy. The raw scores of commonly used scoring functions obtained for the different binding sites are cor. for the different water occupancies by including semi-empirical terms that were obtained from rigorous free energy calcns. Thus the ligands scores are rendered comparable and indipendent of the water occupancy in the binding site used in each case. Results are presented for Pantothenate Synthetase a protein involved in the metab. of M. Tubercolosis. We discuss details of our methodol. and their effect on the obtained correlations between calcd. and exptl. relative binding affinities.

Lead optimization in the course of rational drug design requires determination of binding affinities with high accuracy. In recent years a number of algorithms, most notably Molecular Mechanics Poisson Boltzmann Surface Area (MM/PBSA) and Linear Interaction Energies (LIE), have become increasingly popular as methods that can possibly achieve this aim, providing thereby an attractive alternative to computationally more intensive methods, such as thermodynamic integration. (TI) We investigate the predictive power and generality of MM/PBSA and LIE by applying them to estimate binding affinities of ligands of Hsp90, a protein that has emerged as a promising oncology target. An extensive and well curated test set was generated from in-house data. It includes three con-generic series covering a wide range of affinities that were measured in a consistent fashion, and the X-ray structures for all the receptor/ligand complexes under study. In some cases we can discriminate between weak and strong binders, but find that, for both methods, the optimal parameters can depend sensitively on the tested compound series. We also present a number of results obtained with TI for the same target. We find that, for the target and the ligands considered here, this methodology provides the accuracy required for lead optimization - provided electrostatic long range interactions are taken into account in a consistent way. Our results will contribute to clarify the strength and weaknesses of the various methods both in terms of their underlying principles and their practical applicability.

There exists a broad consensus in the tubercolosis research community that the key to shortening the duration of drug therapy without an increase in relapse rates lies in the specific targeting of proteins that are essential to the non-replicating persistence of Mycobacterium tuberculosis. One such target is pantothenate synthetase (PS). It has been shown that PS is crucial for persistent growth in mice and is over-expressed in in vitro models of NRP. Here we present results of a hierarchical screening approach for searching potent inhibitors of PS. Docking and consensus scoring is followed by application of free energy based methods. Particular attention is paid to the role of crystal waters and receptor flexibility by extending the docking stage to include multiple binding site models. We present first results of these efforts and discuss implications for further optimization of PS inhibitors

abstract

Agglomeration can have a crucial impact on the yield of crystallisation processes and the product quality. In this thesis molecular scale modelling is used to gain insights into the mechanism of crystal agglomeration and factors that determine its progress. The first study analyses the use of the attachment energy model to predict which surfaces of a crystal would be observed and hence the crystal morphology. It is shown that comparatively simple model potentials suffice to provide reasonably accurate morphology predictions within the limitations of the neglect of solvent. A classical empirical force field for potash alum (KAl(SO4 )2 12 H2O) is developed. After having established its capability to reproduce experimentally determined properties of the crystal bulk and the solution, different potash alum crystal faces in contact with aqueous solution are modelled via Molecular Dynamics simulations. A range of different methods of modelling polar crystal surfaces, including a novel one, are investigated. The results are used to rationalise experimental results quantifying the agglomerative strength of PA crystallites as a function of super-saturation and the structure of the crystal faces. Common models for the theoretical prediction of crystal agglomeration include an efficiency parameter which is essentially a material property and a functional of the aver- age force between two particles in solution. A set of Molecular Dynamics simulations of potassium chloride nano-crystallites in aqueous KCl solution is performed in order to es- tablish whether it is possible to obtain reproduceable forces using an explicit water model and an extended system geometry to maintain constant chemical potential of the solution in between two crystal surfaces and a bulk phase. Although the results highlight some interesting aspects, and can give qualitative explanations of agglomeration tendencies, quantitative predictions of agglomeration will require further research.

Separation Shifted Scaling, a computational method coping with the Van der Waal's endpoint problem in free energy simulations, was implemented in the MD-package CHARMM. Test-calculations were carried out for various systems to verify the correctness of the computer code. The free energies of hydration of neon, as well as of water, were calculated using thermodynamic integration in combination with the separation shifted potential for Van der Waal's and Coulomb interactions. The method proved to be very efficient and yielded results with good precision. The calculations for water were repeated for different sizes of the simulation system and cutoff radii in combination with different methods to handle the long range interactions. Results obtained with Ewald-summation, several tapering functions and a generalized reaction field potential were compared. Results obtained with Ewald summation were independent of system size. Thus they were used as reference values. The other free energy results turned out to depend sensitively on the cutoff-radius and on the method treatment of the long range interactions. A cutoff larger than 12 Angstrom has to be used to get results independent of the choice of the tapering function. Only group based switching of the Coulomb interactions with electrically neutral groups yielded results independent of the cutoff-radius.