Domenico Gadaleta

Applicability Domain for QSAR models: where theory meets reality

Quantitative Structure-Activity Relationships are widely acknowledged predictive methods employed, for years, in organic and medicinal chemistry. More recently, they have assumed a central role also in the context of the explorative toxicology for the protection of environment and human health. However, their real-life application has not been always enthusiastically welcomed, being often retrospectively used and, thus, of limited importance for prospective goals. The need of making more trustable predictions has thus addressed studies on the so-called Applicability Domain, which represents the chemical space from which a model is derived and where a prediction is considered to be reliable. In the present study, the authors survey a number of approaches used to build the Applicability Domain. In particular, they will focus on strategies based on: a) physico-chemical, b) structural and c) response domains. Moreover, some examples integrating different strategies will be also discussed to meet the needs of both model developers and downstream users. KeyWoRDS Applicability Domain, Interpolation Space, QSAR, Similarity, Structural Fragments

REACH and in silico methods: an attractive opportunity for medicinal chemists.

REACH, the most ambitious chemical legislation in the world, provides unprecedented opportunities for medicinal chemists. Companies must report (eco)toxicological information about their chemicals, disseminated to the public domain by the European Chemicals Agency after their registration. The availability of this wealth of new toxicological data, together with the promotion of REACH of in silico methods, appoints medicinal chemists to a leading role in the regulatory hazard assessment process. In fact, Quantitative Structure-Activity Relation ship (QSAR) models and predictive toxicology have been applied in drug design and development for decades. Here, we discuss toxicological endpoints and areas where further development is needed to provide an enlightened appraisal of this attractive new opportunity.

Structure-Based Design and Optimization of Multitarget-Directed 2H-Chromen-2-one Derivatives as Potent Inhibitors of Monoamine Oxidase B and Cholinesterases

The multifactorial nature of Alzheimers disease calls for the development of multitarget agents addressing key pathogenic processes. To this end, by following a docking-assisted hybridization strategy, a number of aminocoumarins were designed, prepared, and tested as monoamine oxidases (MAOs) and acetyl- and butyryl-cholinesterase (AChE and BChE) inhibitors. Highly flexible N-benzyl-N-alkyloxy coumarins 2-12 showed good inhibitory activities at MAO-B, AChE, and BChE but low selectivity. More rigid inhibitors, bearing meta- and para-xylyl linkers, displayed good inhibitory activities and high MAO-B selectivity. Compounds 21, 24, 37, and 39, the last two featuring an improved hydrophilic/lipophilic balance, exhibited excellent activity profiles with nanomolar inhibitory potency toward hMAO-B, high hMAO-B over hMAO-A selectivity and submicromolar potency at hAChE. Cell-based assays of BBB permeation, neurotoxicity, and neuroprotection supported the potential of compound 37 as a BBB-permeant neuroprotective agent against H2O2-induced oxidative stress with poor interaction as P-gp substrate and very low cytotoxicity.

An alternative QSAR-based approach for predicting the bioconcentration factor for regulatory purposes

The REACH (Registration, Evaluation, Authorization and restriction of Chemicals) and BPR (Biocide Product Regulation) regulations strongly promote the use of non-animal testing techniques to evaluate chemical risk. This has renewed the interest towards alternative methods such as QSAR in the regulatory context. The assessment of Bioconcentration Factor (BCF) required by these regulations is expensive, in terms of costs, time, and laboratory animal sacrifices. Herein, we present QSAR models based on the ANTARES dataset, which is a large collection of known and verified experimental BCF data. Among the models developed, the best results were obtained from a nine-descriptor highly predictive model. This model was derived from a training set of 608 chemicals and challenged against a validation and blind set containing 152 and 76 chemicals. The models robustness was further controlled through several validation strategies and the implementation of a multi-step approach for the applicability domain. Suitable safety margins were used to increase sensitivity. The easy interpretability of the model is ensured by the use of meaningful biokinetics descriptors. The satisfactory predictive power for external compounds suggests that the new models could represent a reliable alternative to the in vivo assay, helping the registrants to fulfill regulatory requirements in compliance with the ethical and economic necessity to reduce animal testing.

Fine molecular tuning at position 4 of 2H-chromen-2-one derivatives in the search of potent and selective monoamine oxidase B inhibitors

The effects on the inhibition potencies of monoamine oxidase isoforms A (MAO-A) and B (MAO-B) depending upon changes in the physicochemical properties (size, shape, H-bonding, lipophilicity, etc.) of substituents at the C4 position of 2H-chromen-2-one derivatives were extensively investigated, and the results significantly added to our knowledge on this class of MAO inhibitors. All the 67 examined compounds showed high MAO-B selectivity, some of them achieving potency in the low nanomolar range. In particular, the 7-metachlorobenzyloxy-4-oxyacetamido-2H-chromen-2-one (entry 62) showed single digit nanomolar MAO-B potency (IC₅₀ = 3.1 nM) and high selectivity over the MAO-A isoform (selectivity ratio = 7244). The great variety of the investigated substituents at C4 of the 2H-chromen-2-one nucleus, combined with binding models generated from docking studies carried out on selected compounds, allowed us to shed light on the main molecular requirements for potent and selective MAO-B inhibition, highlighting the dominant role of the steric effects. Interestingly, many of the designed substituents could be metabolically related to each other (e.g., CH₃/CH₂OH/CHO/COOH; NH₂/NHCH₃, NHAc), and therefore the results obtained may help in predicting the in vivo activity of some putative metabolites of lead MAO-B inhibitors.

A k-NN algorithm for predicting oral sub-chronic toxicity in the rat

Repeated dose toxicity is of the utmost importance to characterize the toxicological profile of a chemical after repeated administration. Its evaluation refers to the Lowest-Observed-(Adverse)-Effect-Level (LO(A)EL) explicitly requested in several regulatory contexts, such as REACH and EC Regulation 1223/2009 on cosmetic products. So far in vivo tests have been the sole viable option to assess repeated dose toxicity. We report a customized k-Nearest Neighbors approach for predicting sub-chronic oral toxicity in rats. A training set of 254 chemicals was used to derive models whose robustness was challenged through leave-one-out cross-validation. Their predictive power was evaluated on an external dataset comprising 179 chemicals. Despite the intrinsically heterogeneous nature of the data, our models give promising results, with q²≥0.632 and external r²≥0.543. The confidence in prediction was ensured by implementing restrictive user-adjustable rules excluding suspicious chemicals irrespective of the goodness in their prediction. Comparison with the very few LO(A)EL predictive models in the literature indicates that the results of the present analysis can be valuable in prioritizing the safety assessment of chemicals and thus making safe decisions and justifying waiving animal tests according to current regulations concerning chemical safety.

Novel chemotypes targeting tubulin at the colchicine binding site and unbiasing P-glycoprotein

Retrospective validation studies carried out on three benchmark databases containing a small fraction (that is 2.80%) of known tubulin binders permitted us to develop a computational platform very effective in selecting easier manageable subsets showing by far higher percentages of actives (about 25%). These studies relied on the hierarchical application of multilayer in silico screenings employing filters implying molecular shape similarity; a structure-based pharmacophore model and molecular docking campaigns. Building on this validated approach, we performed intensive prospective studies to screen a large chemical collection, including up to 3.7 millions of commercial compounds, to across an unexplored and patent space in the search of novel colchicine binding site inhibitors. Our investigation was successful in identifying a pool of 31 initial hits showing new molecular scaffolds (such as 4,5-dihydro-1H-pyrrolo[3,4-c]pyrazol-6-one and pyrazolo[1,5-a]pyrimidine). This panel of new hits resulted antiproliferative activity in the low μM range towards MCF-7 human breast cancer, HepG2 human liver cancer, HeLa human ovarian cancer and SHSY5Y human glioblastoma cell lines as well as interesting concentration-dependent inhibition of tubulin polymerization assessed through fluorescence polymerization assays. Unlike typical tubulin inhibitors, a satisfactorily low sensitivity towards P-gp was also measured in bi-directional transport studies across MDCKII-MDR1 cells for a selected subset of seven compounds.

Identification of structural alerts for liver and kidney toxicity using repeated dose toxicity data

BackgroundThe potential for a compound to cause hepatotoxicity and nephrotoxicity is a matter of extreme interest for human health risk assessment. To assess liver and kidney toxicity, repeated-dose toxicity (RDT) studies are conducted mainly on rodents. However, these tests are expensive, time-consuming and require large numbers of animals. For early toxicity screening, in silico models can be applied, reducing the costs, time and animals used. Among in silico approaches, structure–activity relationship (SAR) methods, based on the identification of chemical substructures (structural alerts, SAs) related to a particular activity (toxicity), are widely employed.ResultsWe identified and evaluated some SAs related to liver and kidney toxicity, using RDT data on rats taken from the hazard evaluation support system (HESS) database. We considered only SAs that gave the best percentages of true positives (TP).ConclusionsIt was not possible to assign an unambiguous mode of action for all the SAs, but a mechanistic explanation is provided for some of them. Such achievements may help in the early identification of liver and renal toxicity of substances.

CORAL: model for no observed adverse effect level (NOAEL)

The in vivo repeated dose toxicity (RDT) test is intended to provide information on the possible risk caused by repeated exposure to a substance over a limited period of time. The measure of the RDT is the no observed adverse effect level (NOAEL) that is the dose at which no effects are observed, i.e., this endpoint indicates the safety level for a substance. The need to replace in vivo tests, as required by some European Regulations (registration, evaluation authorization and restriction of chemicals) is leading to the searching for reliable alternative methods such as quantitative structure–activity relationships (QSAR). Considering the complexity of the RDT endpoint, for which data quality is limited and depends anyway on the study design, the development of QSAR for this endpoint is an attractive task. Starting from a dataset of 140 organic compounds with NOAEL values related to oral short term toxicity in rats, we developed a QSAR model based on optimal descriptors calculated with simplified molecular input-line entry systems and the graph of atomic orbitals by the Monte Carlo method, using CORAL software. Three different splits into the training, calibration, and validation sets are studied. The mechanistic interpretation of these models in terms of molecular fragment with positive or negative contributions to the endpoint is discussed. The probabilistic definition for the domain of applicability is suggested.

A knowledge-based expert rule system for predicting mutagenicity (Ames test) of aromatic amines and azo compounds.

Cancer is one of the main causes of death in Western countries, and a major issue for human health. Prolonged exposure to a number of chemicals was observed to be one of the primary causes of cancer in occupationally exposed persons. Thus, the development of tools for identifying hazardous chemicals and the increase of mechanistic understanding of their toxicity is a major goal for scientific research. We constructed a new knowledge-based expert system accounting the effect of different substituents for the prediction of mutagenicity (Ames test) of aromatic amines, a class of compounds of major concern because of their widespread application in industry. The herein presented model implements a series of user-defined structural rules extracted from a database of 616 primary aromatic amines, with their Ames test outcomes, aimed at identifying mutagenic and non-mutagenic chemicals. The chemical rationale behind such rules is discussed. Besides assessing the models ability to correctly classify aromatic amines, its predictivity was further evaluated on a second database of 354 azo dyes, another class of chemicals of major concern, whose toxicity has been predicted on the basis of the toxicity of aromatic amines potentially generated from the metabolic reduction of the azo bond. Good performance in classification on both the amine (MCC, Matthews Correlation Coefficient=0.743) and the azo dye (MCC=0.584) datasets confirmed the predictive power of the model, and its suitability for use on a wide range of chemicals. Finally, the model was compared with a series of well-known mutagenicity predicting software. The good performance of our model compared with other mutagenicity models, especially in predicting azo dyes, confirmed the usefulness of this expert system as a reliable support to in vitro mutagenicity assays for screening and prioritization purposes. The model has been fully implemented as a KNIME workflow and is freely available for downstream users.