Current trends in lead discovery: are we looking for the appropriate properties?

Drug-like Properties and Fraction Lipophilicity Index as a combined metric

Fraction Lipophicity Index (FLI) has been developed as a composite drug-like metric combining log P and log D in a weighted manner. In the present study, an extended data set confirmed the previously established drug-like FLI range 0-8 using two calculation systems for log P/log D assessment, the freeware MedChem Designer and ClogP. The dataset was split into two classes according to the percentage of fraction absorbed (%FA) - class 1 including drugs with high to medium absorption levels and class 2 including poorly absorbed drugs. The FLI and FLI-C (ClogP based FLI) drug-like range covers 92 % and 91 % of class 1 drugs, respectively. Using MlogP, a narrower drug-like FLI-M range 0-7 was established, covering 91 % of class 1 drugs. The dependence of the degree of ionization to intrinsic lipophilicity within the FLI (FLI-C, FLI-M) drug-like range as well as the inter-relation between the other Ro5 properties (Mw, HD, HA) was explored to define drug-like / non-drug-like combinations as a safer alternative to single properties for drug candidates' prioritization. In this sense, we propose a combined metric of Mw and the number of polar atoms (Mw/NO) to account for both size and polarity. Setting the value 50 as cutoff, a distinct differentiation between class 1 and class 2 drugs was obtained with Mw/NO>50 for more than 70 % of class 1 drugs, while the opposite was observed for class 2 drugs.

MRlogP: Transfer Learning Enables Accurate logP Prediction Using Small Experimental Training Datasets

Small molecule lipophilicity is often included in generalized rules for medicinal chemistry. These rules aim to reduce time, effort, costs, and attrition rates in drug discovery, allowing the rejection or prioritization of compounds without the need for synthesis and testing. The availability of high quality, abundant training data for machine learning methods can be a major limiting factor in building effective property predictors. We utilize transfer learning techniques to get around this problem, first learning on a large amount of low accuracy predicted logP values before finally tuning our model using a small, accurate dataset of 244 druglike compounds to create MRlogP, a neural network-based predictor of logP capable of outperforming state of the art freely available logP prediction methods for druglike small molecules. MRlogP achieves an average root mean squared error of 0.988 and 0.715 against druglike molecules from Reaxys and PHYSPROP. We have made the trained neural network predictor and all associated code for descriptor generation freely available. In addition, MRlogP may be used online via a web interface.

In Silico Resources of Drug-Likeness as a Mirror: What Are We Lacking in Pesticide-Likeness?

Unfavorable bioavailability is an important aspect underlying the failure of drug candidates. Computational approaches for evaluating drug-likeness can minimize these risks. Over the past decades, computational approaches for evaluating drug-likeness have sped up the process of drug development and were also quickly derived to pesticide-likeness. As a result of many critical differences between drugs and pesticides, many kinds of methods for drug-likeness cannot be used for pesticide-likeness. Therefore, it is crucial to comprehensively compare and analyze the differences between drug-likeness and pesticide-likeness, which may provide a basis for solving the problems encountered during the evaluation of pesticide-likeness. Here, we systematically collected the recent advances of drug-likeness and pesticide-likeness and compared their characteristics. We also evaluated the current lack of studies on pesticide-likeness, the molecular descriptors and parameters adopted, the pesticide-likeness model on pesticide target organisms, and comprehensive analysis tools. This work may guide researchers to use appropriate methods for developing pesticide-likeness models. It may also aid non-specialists to understand some important concepts in drug-likeness and pesticide-likeness.

From Antibacterial to Antitumour Agents: A Brief Review on The Chemical and Medicinal Aspects of Sulfonamides

Sulfonamides have been in clinical use for many years, and the development of bioactive substances containing the sulfonamide subunit has grown steadily in view of their important biological properties such as antibacterial, antifungal, antiparasitic, antioxidant, and antitumour properties. This review addresses the medicinal chemistry aspects of sulfonamides; covering their discovery, the structure- activity relationship and the mechanism of action of the antibacterial sulfonamide class, as well as the physico-chemical and pharmacological properties associated with this class. It also provides an overview of the various biological activities inherent to sulfonamides, reporting research that emphasises the importance of this group in the planning and development of bioactive substances, with a special focus on potential antitumour properties. The synthesis of sulfonamides is considered to be simple and provides a diversity of derivatives from a wide variety of amines and sulfonyl chlorides. The sulfonamide group is a non-classical bioisostere of carboxyl groups, phenolic hydroxyl groups and amide groups. This review highlights that most of the bioactive substances have the sulfonamide group, or a related group such as sulfonylurea, in an orientation towards other functional groups. This structural characteristic was observed in molecules with distinct antibacterial activities, demonstrating a clear structure-activity relationship of sulfonamides. This short review sought to contextualise the discovery of classic antibacterial sulfonamides and their physico-chemical and pharmacological properties. The importance of the sulfonamide subunit in Medicinal Chemistry has been highlighted and emphasised, in order to promote its inclusion in the planning and synthesis of future drugs.

Fraction Lipophilicity Index (FLI). A drug-like metric for orally administered ionizable drugs

Fraction Lipophilicity Index (FLI) was developed as a metric for assessing drug likeness of ionizable oral drugs. Considering that both log P and log D have distinct roles in drug action, the metric FLI allocates lipophilicity to a pH dependent neutral fraction of the molecule and is a weighted combination of log P and log D. It is expressed by equation: FLI = 2 x log P-│log D│. A dataset of 368 basic and acidic drugs was analyzed. Based on available % absorption data, drugs were classified into class 1 (268 drugs) and class 2 (100 drugs). The freeware MedChem Designer was used for log P and log D calculations at pH 7.4 and pH 5.5 for acids. Based on class 1, a drug-like FLI range 0-8 was defined. FLI distribution for class 2 is shifted towards significantly lower values. Comparison of FLI with rule of 5 (Ro5) shows that it leads to fewer values outside the established range than the corresponding log P violations for class 1. For class 2 it gives more alerts than Ro5 and can be considered complementary to Ro5, while it also sets lower limits to discriminate drugs with poor absorption.

How to Prepare a Compound Collection Prior to Virtual Screening

Virtual screening is a well-established technique that has proven to be successful in the identification of novel biologically active molecules, including drug repurposing. Whether for ligand-based or for structure-based virtual screening, a chemical collection needs to be properly processed prior to in silico evaluation. Here we describe our step-by-step procedure for handling very large collections (up to billions) of compounds prior to virtual screening.

Chromatographic and Computational Studies of Molecular Lipophilicity and Drug-likeness for few 2-Thioxo-1,3-Thiazolidin-4-one Derivatives and their Analogs

Hydrophobicity of the eight 2-thioxo-1,3-thiazolidin-4-one derivatives was determined experimentally by thin-layer chromatography and predicted by means of commercially available programmers. RM values were determined by reversed-phase thin-layer chromatography with using acetonitrile-water, methanol-water, acetone-water, propan-2-ol-water or 1,4-dioxane-water and compared with logP values calculated by using computer programs: HyperChem 8.0.10, Virtual Chemical Calculation Laboratory, ACD/LogP. The drug-likeness has been calculated using Molinspiration. All the heterocycles were found to obey Lipinski's rule of 5 for an orally active drug.

Structure based drug discovery for designing leads for the non-toxic metabolic targets in multi drug resistant Mycobacterium tuberculosis

Background

The problem of drug resistance and bacterial persistence in tuberculosis is a cause of global alarm. Although, the UN’s Sustainable Development Goals for 2030 has targeted a Tb free world, the treatment gap exists and only a few new drug candidates are in the pipeline. In spite of large information from medicinal chemistry to ‘omics’ data, there has been a little effort from pharmaceutical companies to generate pipelines for the development of novel drug candidates against the multi drug resistant Mycobacterium tuberculosis.

Methods

In the present study, we describe an integrated methodology; utilizing systems level information to optimize ligand selection to lower the failure rates at the pre-clinical and clinical levels. In the present study, metabolic targets (Rv2763c, Rv3247c, Rv1094, Rv3607c, Rv3048c, Rv2965c, Rv2361c, Rv0865, Rv0321, Rv0098, Rv0390, Rv3588c, Rv2244, Rv2465c and Rv2607) in M. tuberculosis, identified using our previous Systems Biology and data-intensive genome level analysis, have been used to design potential lead molecules, which are likely to be non-toxic. Various in silico drug discovery tools have been utilized to generate small molecular leads for each of the 15 targets with available crystal structures.

Results

The present study resulted in identification of 20 novel lead molecules including 4 FDA approved drugs (droxidropa, tetroxoprim, domperidone and nemonapride) which can be further taken for drug repurposing. This comprehensive integrated methodology, with both experimental and in silico approaches, has the potential to not only tackle the MDR form of Mtb but also the most important persister population of the bacterium, with a potential to reduce the failures in the Tb drug discovery.

Conclusion

We propose an integrated approach of systems and structural biology for identifying targets that address the high attrition rate issue in lead identification and drug development We expect that this system level analysis will be applicable for identification of drug candidates to other pathogenic organisms as well.

Electronic supplementary material

The online version of this article (10.1186/s12967-017-1363-9) contains supplementary material, which is available to authorized users.

Why and how have drug discovery strategies in pharma changed? What are the new mindsets?

In the pharmaceutical industry the long-term challenge of drug innovation is the key phrase throughout R&D that refers to increasing the output of original drug candidate molecules. To increase R&D productivity, implementation of new and strategic R&D orientations to develop new approaches or systems to identify hits and leads efficiently has taken place and enabled all scientists working in the drug discovery domain to develop innovative medicines for the 21st century.

Influence of n-octanol in mobile phase on QSRRs of lipophilicity and retention mechanism of acidic and basic compounds in RP-HPLC

Effect ofn-octanol on retention behavior, mechanism and QSRRs of ionizable compounds in RP-HPLC.

The impact of physicochemical and molecular properties in drug design: navigation in the "drug-like" chemical space

Physicochemical and molecular properties influence both pharmacokinetic and pharmacodynamic process, as well as drug safety, often in a conflicting way. In this aspect the current trend in drug discovery is to consider ADME (T) properties in parallel with target affinity. The concept of "drug-likeness" defines acceptable boundaries of fundamental properties formulated as simple rules of thumb, in order to aid the medicinal chemist to prioritize drug candidates. Special attention is given to lipophilicity and molecular weight, since there is a tendency for those parameters to increase in regard to complex compounds generated by new technologies, with potential consequences in bioavailability, while high lipophilicity is also associated with undesired effects. Such rules have the advantage to be very simple and are easy to interpret; however their drawback is that they do not take into consideration uncertainties in measurements and calculations as well as the receptor requirements. The case of PPARs, a nuclear receptor family, is discussed in detail in regard to the chemical space covered by the ligands, focusing on the high demands of the ligand binding domain in both lipophilicity and molecular size. Such paradigms indicate that it would be more appropriate to adapt drug-like properties according to specific drug discovery projects.

Novel tactics for designing water-soluble molecules in drug discovery

INTRODUCTION

Physiochemical drug properties, such as aqueous solubility are considered to be a major factor in determining the ultimate success or failure of experimental agents. Solubility is important because it determines the maximum dose which can be taken up. As the size and hydrophobicity of drug candidates has increased over the years, poor solubility has become a more prevalent issue. Recent examples from the literature show that an improved understanding of the relationship between molecular structure and solubility allows this issue to be approached using rational design.

AREAS COVERED

This review provides selected examples from the recent drug discovery literature that demonstrate various tactics, which have been applied successfully towards improving drug solubility. The examples that were selected demonstrate the underlying principles behind aqueous solubility, such as hydrophobicity and crystalline stability.

EXPERT OPINION

From a strategic point of view, improving the solubility of a compound should be straightforward because it can be accomplished by simply reducing hydrophobicity or crystalline stability. However, the structural elements and physical properties which control solubility also influence potency, pharmacokinetics and toxicity. Furthermore, there are practical aspects such as the quantity and quality of solubility-related data, which hamper the development of structure-solubility relationships. Given that poor aqueous solubility remains a primary issue in drug discovery, there is a continuous need for novel methods to overcome it.

Virtualizing the p-ANAPL library: a step towards drug discovery from African medicinal plants

Background

Natural products play a key role in drug discovery programs, both serving as drugs and as templates for the synthesis of drugs, even though the quantities and availabilities of samples for screening are often limitted.

Experimental approach

A current collection of physical samples of > 500 compound derived from African medicinal plants aimed at screening for drug discovery has been made by donations from several researchers from across the continent to be directly available for drug discovery programs. A virtual library of 3D structures of compounds has been generated and Lipinski’s “Rule of Five” has been used to evaluate likely oral availability of the samples.

Results

A majority of the compound samples are made of flavonoids and about two thirds (2/3) are compliant to the “Rule of Five”. The pharmacological profiles of thirty six (36) selected compounds in the collection have been discussed.

Conclusions and implications

The p-ANAPL library is the largest physical collection of natural products derived from African medicinal plants directly available for screening purposes. The virtual library is also available and could be employed in virtual screening campaigns.

How "drug-like" are naturally occurring anti-cancer compounds?

We attempt to evaluate the "drug-likeness" of a collection of ∼1500 natural products, exhibiting in vitro or in vivo activities against cancers of various forms, by using a set of calculated molecular descriptors. Compliance to Lipinski's "Rule of Five" and Jorgensen's "Rule of Three" have been used to assess oral availability, by making use of popular parameters like molecular weights, predicted lipophilicities, number of hydrogen bond donors/acceptors, predicted aqueous solubilities, number of primary metabolites and Caco-2 permeabilities. Meanwhile 24 descriptors have been used to predict properties related to the absorption, distribution, metabolism, elimination, and toxicity (ADMET). The ADMET profiles of the anticancer natural products have been analyzed in comparision with the range of properties for 95 % of known drugs. Our results show that the computed parameters fall within the recommended range for about 42 % of the studied compounds, while respectively 63 % and 69 % of the corresponding 'drug-like' and 'lead-like' subsets had properties predicted to fall within the recommended range for 95 % of known drugs. The aim of giving a picture of how drug-like they are and bring out the need to return to natural sources in searching for anticancer lead compounds.

The uniqueness and therapeutic value of natural products from West African medicinal plants, part II: terpenoids, geographical distribution and drug discovery

In this review series, an attempt has been made to give indepth coverage of natural products derived from West African medicinal plants with diverse biological activities.

New antimalarial hits from Dacryodes edulis (Burseraceae)--part I: isolation, in vitro activity, in silico "drug-likeness" and pharmacokinetic profiles

The aims of the present study were to identify the compounds responsible for the anti-malarial activity of Dacryoedes edulis (Burseraceae) and to investigate their suitability as leads for the treatment of drug resistant malaria. Five compounds were isolated from ethyl acetate and hexane extracts of D. edulis stem bark and tested against 3D7 (chloroquine-susceptible) and Dd2 (multidrug-resistant) strains of Plasmodium falciparum, using the parasite lactate dehydrogenase method. Cytotoxicity studies were carried out on LLC-MK2 monkey kidney epithelial cell-line. In silico analysis was conducted by calculating molecular descriptors using the MOE software running on a Linux workstation. The “drug-likeness” of the isolated compounds was assessed using Lipinski criteria, from computed molecular properties of the geometry optimized structures. Computed descriptors often used to predict absorption, distribution, metabolism, elimination and toxicity (ADMET) were used to assess the pharmacokinetic profiles of the isolated compounds. Antiplasmodial activity was demonstrated for the first time in five major natural products previously identified in D. edulis, but not tested against malaria parasites. The most active compound identified was termed DES4. It had IC₅₀ values of 0.37 and 0.55 µg/mL, against 3D7 and Dd2 respectively. In addition, this compound was shown to act in synergy with quinine, satisfied all criteria of “Drug-likeness” and showed considerable probability of providing an antimalarial lead. The remaining four compounds also showed antiplasmodial activity, but were less effective than DES4. None of the tested compounds was cytotoxicity against LLC-MK2 cells, suggesting their selective activities on malaria parasites. Based on the high in vitro activity, low toxicity and predicted “Drug-likeness” DES4 merits further investigation as a possible drug lead for the treatment of malaria.

AfroDb: a select highly potent and diverse natural product library from African medicinal plants

Computer-aided drug design (CADD) often involves virtual screening (VS) of large compound datasets and the availability of such is vital for drug discovery protocols. We assess the bioactivity and "drug-likeness" of a relatively small but structurally diverse dataset (containing >1,000 compounds) from African medicinal plants, which have been tested and proven a wide range of biological activities. The geographical regions of collection of the medicinal plants cover the entire continent of Africa, based on data from literature sources and information from traditional healers. For each isolated compound, the three dimensional (3D) structure has been used to calculate physico-chemical properties used in the prediction of oral bioavailability on the basis of Lipinski's "Rule of Five". A comparative analysis has been carried out with the "drug-like", "lead-like", and "fragment-like" subsets, as well as with the Dictionary of Natural Products. A diversity analysis has been carried out in comparison with the ChemBridge diverse database. Furthermore, descriptors related to absorption, distribution, metabolism, excretion and toxicity (ADMET) have been used to predict the pharmacokinetic profile of the compounds within the dataset. Our results prove that drug discovery, beginning with natural products from the African flora, could be highly promising. The 3D structures are available and could be useful for virtual screening and natural product lead generation programs.

Targeting UDP-galactopyranose mutases from eukaryotic human pathogens

UDP-Galactopyranose mutase (UGM) is a unique flavin-dependent enzyme that catalyzes the conversion of UDP-galactopyranose(UDP-Galp) to UDP-galactofuranose (UDP-Galf). The product of this reaction is the precursor to Galf, a major component of the cell wall and of cell surface glycoproteins and glycolipids in many eukaryotic and prokaryotic human pathogens. The function of UGM is important in the virulence of fungi, parasites, and bacteria. Its role in virulence and its absence in humans suggest that UGM is an ideal drug target. Significant structural and mechanistic information has been accumulated on the prokaryotic UGMs; however, in the past few years the research interest has shifted to UGMs from eukaryotic human pathogens such as fungi and protozoan parasites. It has become clear that UGMs from prokaryotic and eukaryotic organisms have different structural and mechanistic features. The amino acid sequence identity between these two classes of enzymes is low, resulting in differences in oligomeric states, substrate binding, active site flexibility, and interaction with redox partners. However, the unique role of the flavin cofactor in catalysis is conserved among this enzyme family. In this review, recent findings on eukaryotic UGMs are discussed and presented in comparison with prokaryotic UGMs.

The Collaborative Drug Discovery (CDD) database

The broad goals of Collaborative Drug Discovery (CDD) are to enable a collaborative "cloud-based" tool to be used to bring together neglected disease researchers and other researchers from usually separate areas, to collaborate and to share compounds and drug discovery data in the research community, which will ultimately result in long-term improvements in the research enterprise and health care delivery. This chapter briefly introduces CDD software and describes applications in antimalarial and tuberculosis research.

Compound collection preparation for virtual screening

Virtual screening is an established technique that has successfully been deployed in the identification of novel biologically active molecules. Whether for ligand-based or for structure-based virtual screening, a chemical collection needs to be properly processed prior to in silico evaluation. Here we describe our step-by-step procedure for handling large collections of compounds prior to virtual screening.

Assessing the pharmacokinetic profile of the CamMedNP natural products database: an in silico approach

Background

Drug metabolism and pharmacokinetic (DMPK) assessment has come to occupy a place of interest during the early stages of drug discovery today. Computer-based methods are slowly gaining ground in this area and are often used as initial tools to eliminate compounds likely to present uninteresting pharmacokinetic profiles and unacceptable levels of toxicity from the list of potential drug candidates, hence cutting down the cost of the discovery of a drug.

Results

In the present study, we present an in silico assessment of the DMPK profile of our recently published natural products database of 1,859 unique compounds derived from 224 species of medicinal plants from the Cameroonian forest. In this analysis, we have used 46 computed physico-chemical properties or molecular descriptors to predict the absorption, distribution, metabolism and elimination (ADME) of the compounds. This survey demonstrated that about 50% of the compounds within the Cameroonian medicinal plant and natural products (CamMedNP) database are compliant, having properties which fall within the range of ADME properties of >95% of currently known drugs, while >73% of the compounds have ≤2 violations. Moreover, about 72% of the compounds within the corresponding ‘drug-like’ subset showed compliance.

Conclusions

In addition to the previously verified levels of ‘drug-likeness’ and the diversity and the wide range of measured biological activities, the compounds in the CamMedNP database show interesting DMPK profiles and, hence, could represent an important starting point for hit/lead discovery from medicinal plants in Africa.

Setting expectations in molecular optimizations: Strengths and limitations of commonly used composite parameters

Over the past 15years there have been extensive efforts to understand and reduce the high attrition rates of drug candidates with an increased focus on physicochemical properties. The fruits of this labor have been the generation of numerous efficiency indices, metric-based rules and visualization tools to help guide medicinal chemists in the design of new compounds with more favorable properties. This deluge of information may have had the unintended consequence of further obfuscating molecular optimizations by the inability of these scoring functions, rules and guides to reach a consensus on when a particular transformation is identified as beneficial. In this manuscript, several composite parameters, or efficiency indices, are examined utilizing theoretical and experimental matched molecular pair analyses in order to understand the basis for how each will perform under varying scenarios of molecular optimizations. In contrast to empirically derived composite parameters based on heavy atom count, lipophilic efficiency (LipE) sets consistent expectations regardless of molecular weight or relative potency and can be used to generate consistent expectations for any matched molecular pair.

In silico drug metabolism and pharmacokinetic profiles of natural products from medicinal plants in the Congo basin

Purpose

Drug metabolism and pharmacokinetics (DMPK) assessment has come to occupy a place of interest during the early stages of drug discovery today. The use of computer modelling to predict the DMPK and toxicity properties of a natural product library derived from medicinal plants from Central Africa (named ConMedNP). Material from some of the plant sources are currently employed in African Traditional Medicine.

Methods

Computer-based methods are slowly gaining ground in this area and are often used as preliminary criteria for the elimination of compounds likely to present uninteresting pharmacokinetic profiles and unacceptable levels of toxicity from the list of potential drug candidates, hence cutting down the cost of discovery of a drug.

In the present study, we present an in silico assessment of the DMPK and toxicity profile of a natural product library containing ~3,200 compounds, derived from 379 species of medicinal plants from 10 countries in the Congo Basin forests and savannas, which have been published in the literature. In this analysis, we have used 46 computed physico-chemical properties or molecular descriptors to predict the absorption, distribution, metabolism and elimination and toxicity (ADMET) of the compounds.

Results

This survey demonstrated that about 45% of the compounds within the ConMedNP compound library are compliant, having properties which fall within the range of ADME properties of 95% of currently known drugs, while about 69% of the compounds have ≤ 2 violations. Moreover, about 73% of the compounds within the corresponding “drug-like” subset showed compliance.

Conclusions

In addition to the verified levels of “drug-likeness”, diversity and the wide range of measured biological activities, the compounds from medicinal plants in Central Africa show interesting DMPK profiles and hence could represent an important starting point for hit/lead discovery.

Electronic supplementary material

The online version of this article (doi:10.1186/2193-9616-1-12) contains supplementary material, which is available to authorized users.

An in silico evaluation of the ADMET profile of the StreptomeDB database

Background

Computer-aided drug design (CADD) often involves virtual screening (VS) of large compound datasets and the availability of such is vital for drug discovery protocols. This paper presents an assessment of the “drug-likeness” and pharmacokinetic profile of > 2,400 compounds of natural origin, currently available in the recently published StreptomeDB database.

Methods

The evaluation of “drug-likeness” was performed on the basis of Lipinski’s “Rule of Five”, while 46 computed physicochemical properties or molecular descriptors were used to predict the absorption, distribution, metabolism, elimination and toxicity (ADMET) of the compounds.

Results

This survey demonstrated that, of the computed molecular descriptors, about 28% of the compounds within the StreptomeDB database were compliant, having properties which fell within the range of ADMET properties of 95% of currently known drugs, while about 44% of the compounds had ≤ 2 violations. Moreover, about 50% of the compounds within the corresponding “drug-like” subset showed compliance, while >83% of the “drug-like” compounds had ≤ 2 violations.

Conclusions

In addition to the previously verified range of measured biological activities, the compounds in the StreptomeDB database show interesting DMPK profiles and hence could represent an important starting point for hit/lead discovery from natural sources. The generated data are available and could be highly useful for natural product lead generation programs.

Electronic supplementary material

The online version of this article (doi:10.1186/2193-1801-2-353) contains supplementary material, which is available to authorized users.

Enhanced hit-to-lead process using bioanalogous lead evolution and chemogenomics: application in designing selective matrix metalloprotease inhibitors

The authors describe an innovative approach for designing novel inhibitors. This approach effectively integrates the emerging chemogenomics concept of target-family-based drug discovery with bioanalogous design strategies, including privileged structures, molecular frameworks as well as bioisosteric and bioanalogous/isofunctional modifications. The authors applied this method in the design of selective inhibitors of matrix metalloproteases (MMPs), also referred to as matrixins, on the basis of a unique analysis of the ligand-target knowledge base, the 'matrixinome'. For this analysis, the authors created an annotated MMP database containing ∼ 300 inhibitors with their published activity profile. The ligand space was then arranged into a lead evolution tree, where the substructural transformations in each virtual step led to marked changes in the activity pattern. This allowed subtype-specific privileged fragments to be extracted as well as modifications, which improve activity and/or selectivity. Furthermore, the compounds with the preferred activity profile were correlated with sequence homology as well as binding site similarity within the target family, thereby leading to the identification of substructural modifications that turn non-selective, biohomologous structures into selective inhibitors. The matrixinomic application of the authors' approach, therefore, provides an example of how the combination of ligand space knowledge with sequence-related data can radically improve the outcome of the lead optimisation process to achieve higher selectivity within a given target family.

Inside the mind of a medicinal chemist: the role of human bias in compound prioritization during drug discovery

Medicinal chemists' "intuition" is critical for success in modern drug discovery. Early in the discovery process, chemists select a subset of compounds for further research, often from many viable candidates. These decisions determine the success of a discovery campaign, and ultimately what kind of drugs are developed and marketed to the public. Surprisingly little is known about the cognitive aspects of chemists' decision-making when they prioritize compounds. We investigate 1) how and to what extent chemists simplify the problem of identifying promising compounds, 2) whether chemists agree with each other about the criteria used for such decisions, and 3) how accurately chemists report the criteria they use for these decisions. Chemists were surveyed and asked to select chemical fragments that they would be willing to develop into a lead compound from a set of ~4,000 available fragments. Based on each chemist's selections, computational classifiers were built to model each chemist's selection strategy. Results suggest that chemists greatly simplified the problem, typically using only 1-2 of many possible parameters when making their selections. Although chemists tended to use the same parameters to select compounds, differing value preferences for these parameters led to an overall lack of consensus in compound selections. Moreover, what little agreement there was among the chemists was largely in what fragments were undesirable. Furthermore, chemists were often unaware of the parameters (such as compound size) which were statistically significant in their selections, and overestimated the number of parameters they employed. A critical evaluation of the problem space faced by medicinal chemists and cognitive models of categorization were especially useful in understanding the low consensus between chemists.

Aqueous solubility: simple predictive methods (in silico, in vitro and bio-relevant approaches)

Aqueous solubility is a key physicochemical attribute required for the characterisation of an active pharmaceutical ingredient (API) during drug discovery and beyond. Furthermore, aqueous solubility is highly important for formulation selection and subsequent development processes. This review provides a summary of simple predictive methods used to assess aqueous solubility as well as an assessment of the more complex in silico methodologies and a review of the recent solubility challenge. In addition, a summary of experimental methods to determine solubility is included, with a discussion of some potential pitfalls.

Diverse inhibitor chemotypes targeting Trypanosoma cruzi CYP51

BACKGROUND

Chagas Disease, a WHO- and NIH-designated neglected tropical disease, is endemic in Latin America and an emerging infection in North America and Europe as a result of population moves. Although a major cause of morbidity and mortality due to heart failure, as well as inflicting a heavy economic burden in affected regions, Chagas Disease elicits scant notice from the pharmaceutical industry because of adverse economic incentives. The discovery and development of new routes to chemotherapy for Chagas Disease is a clear priority.

METHODOLOGY/PRINCIPAL FINDINGS

The similarity between the membrane sterol requirements of pathogenic fungi and those of the parasitic protozoon Trypanosoma cruzi, the causative agent of Chagas human cardiopathy, has led to repurposing anti-fungal azole inhibitors of sterol 14α-demethylase (CYP51) for the treatment of Chagas Disease. To diversify the therapeutic pipeline of anti-Chagasic drug candidates we exploited an approach that included directly probing the T. cruzi CYP51 active site with a library of synthetic small molecules. Target-based high-throughput screening reduced the library of ∼104,000 small molecules to 185 hits with estimated nanomolar K(D) values, while cross-validation against T. cruzi-infected skeletal myoblast cells yielded 57 active hits with EC(50) <10 µM. Two pools of hits partially overlapped. The top hit inhibited T. cruzi with EC(50) of 17 nM and was trypanocidal at 40 nM.

CONCLUSIONS/SIGNIFICANCE

The hits are structurally diverse, demonstrating that CYP51 is a rather permissive enzyme target for small molecules. Cheminformatic analysis of the hits suggests that CYP51 pharmacology is similar to that of other cytochromes P450 therapeutic targets, including thromboxane synthase (CYP5), fatty acid ω-hydroxylases (CYP4), 17α-hydroxylase/17,20-lyase (CYP17) and aromatase (CYP19). Surprisingly, strong similarity is suggested to glutaminyl-peptide cyclotransferase, which is unrelated to CYP51 by sequence or structure. Lead compounds developed by pharmaceutical companies against these targets could also be explored for efficacy against T. cruzi.

Furin inhibitors: importance of the positive formal charge and beyond

Furin is the prototype member of the proprotein convertases superfamily. Proprotein convertases are associated with hormonal response, neural degeneration, viral and bacterial activation, and cancer. Several studies over the last decade have examined small molecules, natural products, peptides and peptide derivatives as furin inhibitors. Currently, subnanomolar inhibition of furin is possible. Herein, we report the analysis of 115 furin inhibitors reported in the literature. Analysis of the physicochemical properties of these compounds highlights the dependence of the inhibitory potency with the total formal charge and also shows how the most potent (peptide-based) furin inhibitors have physicochemical properties similar to drugs. In addition, we report docking studies of 26 furin inhibitors using Glide XP. Inspection of binding interactions shows that the two putative binding modes derived from our study are reasonable. Analysis of the binding modes and protein-ligand interaction fingerprints, used here as postdocking procedure, shows that electrostatic interactions predominate on S1, S2 and S4 subsites but are seldom in S3. Our models also show that the benzimidamide group, present in the most active inhibitors, can be accommodated in the S1 subsite. These results are valuable for the design of new furin inhibitors.

Exploring DNA topoisomerase I ligand space in search of novel anticancer agents

DNA topoisomerase I (Top1) is over-expressed in tumour cells and is an important target in cancer chemotherapy. It relaxes DNA torsional strain generated during DNA processing by introducing transient single-strand breaks and allowing the broken strand to rotate around the intermediate Top1 – DNA covalent complex. This complex can be trapped by a group of anticancer agents interacting with the DNA bases and the enzyme at the cleavage site, preventing further topoisomerase activity. Here we have identified novel Top1 inhibitors as potential anticancer agents by using a combination of structure- and ligand-based molecular modelling methods. Pharmacophore models have been developed based on the molecular characteristics of derivatives of the alkaloid camptothecin (CPT), which represent potent antitumour agents and the main group of Top1 inhibitors. The models generated were used for in silico screening of the National Cancer Institute (NCI, USA) compound database, leading to the identification of a set of structurally diverse molecules. The strategy is validated by the observation that amongst these molecules are several known Top1 inhibitors and agents cytotoxic against human tumour cell lines. The potential of the untested hits to inhibit Top1 activity was further evaluated by docking into the binding site of a Top1 – DNA complex, resulting in a selection of 10 compounds for biological testing. Limited by the compound availability, 7 compounds have been tested in vitro for their Top1 inhibitory activity, 5 of which display mild to moderate Top1 inhibition. A further compound, found by similarity search to the active compounds, also shows mild activity. Although the tested compounds display only low in vitro antitumour activity, our approach has been successful in the identification of structurally novel Top1 inhibitors worthy of further investigation as potential anticancer agents.

BDDCS applied to over 900 drugs

Here, we compile the Biopharmaceutics Drug Disposition Classification System (BDDCS) classification for 927 drugs, which include 30 active metabolites. Of the 897 parent drugs, 78.8% (707) are administered orally. Where the lowest measured solubility is found, this value is reported for 72.7% (513) of these orally administered drugs and a dose number is recorded. The measured values are reported for percent excreted unchanged in urine, LogP, and LogD (7.4) when available. For all 927 compounds, the in silico parameters for predicted Log solubility in water, calculated LogP, polar surface area, and the number of hydrogen bond acceptors and hydrogen bond donors for the active moiety are also provided, thereby allowing comparison analyses for both in silico and experimentally measured values. We discuss the potential use of BDDCS to estimate the disposition characteristics of novel chemicals (new molecular entities) in the early stages of drug discovery and development. Transporter effects in the intestine and the liver are not clinically relevant for BDDCS class 1 drugs, but potentially can have a high impact for class 2 (efflux in the gut, and efflux and uptake in the liver) and class 3 (uptake and efflux in both gut and liver) drugs. A combination of high dose and low solubility is likely to cause BDDCS class 4 to be underpopulated in terms of approved drugs (N = 53 compared with over 200 each in classes 1-3). The influence of several measured and in silico parameters in the process of BDDCS category assignment is discussed in detail.

Finding promiscuous old drugs for new uses

From research published in the last six years we have identified 34 studies that have screened libraries of FDA-approved drugs against various whole cell or target assays. These studies have each identified one or more compounds with a suggested new bioactivity that had not been described previously. We now show that 13 of these drugs were active against more than one additional disease, thereby suggesting a degree of promiscuity. We also show that following compilation of all the studies, 109 molecules were identified by screening in vitro. These molecules appear to be statistically more hydrophobic with a higher molecular weight and AlogP than orphan-designated products with at least one marketing approval for a common disease indication or one marketing approval for a rare disease from the FDA's rare disease research database. Capturing these in vitro data on old drugs for new uses will be important for potential reuse and analysis by others to repurpose or reposition these or other existing drugs. We have created databases which can be searched by the public and envisage that these can be updated as more studies are published.