Current Computer - Aided Drug Design - Volume 7, Issue 3, 2011

Almost all of today's recently released drugs are designed using Computer-Aided Drug Design (CADD) techniques. Among various methods, Structure-Based Design (SBD) and Pharmacophore Modeling are the two most successful methodologies in designing new drug candidates. Whereas SBD requires the receptor structure to be known, there is no such requirement for pharmacophore modeling. This makes pharmacophore modeling a universal tool for CADD. Availability of receptor structure, however, makes the pharmacophore tools even more powerful as demonstrated by the four papers in this issue. The aim of this special issue is to highlight recent developments in this area. The papers focus on success stories, recent developments in pharmacophore-based design, and new approaches and algorithms. The Special Issue starts with an application article on an excellent study by Elizabeth Sobhia and Dara Ajay that utilizes various computational techniques with a receptor-based pharmacophore approach to discover potential inhibitors of PTP-LAR. Methodologically, the authors have employed existing commercial software tools to create receptor-based pharmacophore from an apoenzyme when no bound ligand is present in the crystal structure, which was then used to identify potential ligands that were further validated through docking studies and analyses. From there, we then move to observing how some of these existing tools have been enhanced within the Discovery Studio environment. Jon Sutter and co-worker's paper not only discusses the recent enhancement made to their popular pharmacophore modeling system based on Catalyst, it also describes some of the new tools supporting receptor-based pharmacophores. Weifan Zheng and co-workers describe a new approach in combining the active-site shape and pharmacophore tools to develop more selective pharmacophore models. This approach is an enhancement to the existing commercial tools that involve combined shape-pharmacophore tools......

The design of biological active compounds from the apoenzyme is still a challenging task. Herein a simple yet efficient technique is reported to generate a receptor based pharmacophore solely using a ligand-free protein crystal structure. Human leukocyte antigen-related phosphatase (PTP-LAR) is an apoenzyme and a receptor like transmembrane phosphatase that has emerged as a drug target for diabetes, obesity and cancer. The prior knowledge of the active residues responsible for the mechanism of action of the protein was used to generate the LUDI interaction map. Then, the complement negative image of the binding site was used to generate the pharmacophore features. A unique strategy was followed to design a pharmacophore query maintaining crucial interactions with all the active residues, essential for the enzyme inhibition. The same query was used to screen several databases consisting of the Specs, IBS, MiniMaybridge, NCI and an in-house PTP inhibitor databases. In order to overcome the common bioavailability problem associated with phosphatases, the hits obtained were filtered by Lipinski's Rule of Five, SADMET properties and validated by docking studies in Glide and GOLD. These docking studies not only suggest the essential ligand binding interactions but also the binding patterns necessary for the LAR inhibition. The ligand pharmacophore mapping studies further validated the screened protocol and supported that the final screened molecules, presumably, showed potent inhibitory activity. Subsequently, these molecules were subjected to Derek toxicity predictions and nine new molecules with different scaffold were obtained as non-toxic PTP-LAR inhibitors. The present prospective strategy is a powerful technique to identify potent inhibitors using the protein 3D structure alone and is a valid alternative to other structure-based and random docking approaches.

Generating a pharmacophore is often the first step towards understanding the interactions between a receptor and a ligand and can be pivotal to a successful drug discovery project. The pharmacophore tools at Accelrys have been used to assist in many different projects over the years, such as lead generation, scaffold hopping, mining ligand databases as well as many more. In this article, we will review the pharmacophore tools that have been developed at Accelrys. These will include the often used and well validated ligand based algorithms, HipHop and HypoGen and as well as extensions of these algorithms, HipHopRefine and HypoGenRefine. Recently we also developed new pharmacophore tools in the area of structure based design - deriving pharmacophores from the receptor as well as the receptor-ligand complex - which will also be discussed in this paper.

The intuitive nature of the pharmacophore concept has made it widely accepted by the medicinal chemistry community, evidenced by the past 3 decades of development and application of computerized pharmacophore modeling tools. On the other hand, shape complementarity has been recognized as a critical factor in molecular recognition between drugs and their receptors. Recent development of fast and accurate shape comparison tools has facilitated the wide spread use of shape matching technologies in drug discovery. However, pharmacophore and shape technologies, if used separately, often lead to high false positive rate. Thus, integrating pharmacophore matching and shape matching technologies into one program has the potential to reduce the false positive rates in virtual screening. Other issues of current pharmacophore technologies include sometimes high false negative rate and non-quantitative prediction. In this article, we first focus on a recently implemented method (Shape4) that combines receptor based shape matching and pharmacophore comparison in a single algorithm to create shape pharmacophore models for virtual screening. We also examine a recent example that utilizes multi-complex information to develop receptor-based pharmacophore models that promises to reduce false negative rate. Finally, we review several methods that employ receptor-based pharmacophore map and pharmacophore key descriptors for QSAR modeling. We conclude by emphasizing the concept of receptor-based shape pharmacophore and its roles in future drug discovery.

The question of how and why a small molecule binds to a protein is central to ligand-based drug discovery. The traditional way of approaching these questions is pharmacophore analysis. However, pharmacophores as usually applied lack quantitation and subtlety. An improvement is to consider the electrostatic and steric fields of the ligand directly. Molecular fields provide a rich view of the potential interactions that a molecule can make and can be validated through experimental data on molecular interactions and through quantum mechanics calculations. A technique is presented in this review for comparing molecules using molecular fields and assigning similarity scores. This high information content method can be used to align molecules for SAR analysis, to determine the bioactive conformation from ligand data, and to screen large libraries of compounds for structurally unrelated actives. An extension to allow interactive exploration of chemistry space via bioisostere analysis is also reviewed. Examples from the literature showing the success of these methods are presented, and future directions discussed.

The apparently paradoxical lack of correlation between the huge increase in the discovery of new potential drug targets made possible by the post-genomic sciences and new drugs development has stimulated many different interpretations. Here we illustrate the general principle of redundancy of biological pathways on hand of simplified mathematical approaches applied to different models of biological regulation. The simulation was based on the analysis of the ‘ degree of autonomy’ of network architectures in which the possibility for an external stimulus (e.g. a drug) impinging into a specific node to be sensed by the entire network, and eventually amplified up to a macroscopic consequence, was demonstrated to be limited to strictly linear pathways. The implications of such a result for poly-pharmacology and computational approaches to drug development are described as well.

AMP-activated protein kinase (AMPK) is an important therapeutic target for the potential treatment of metabolic disorders, cardiovascular disease and cancer. Recently, various classes of compounds that activate AMPK by direct or indirect interactions have been reported. The importance of computer-aided drug design approaches in the search for potent activators of AMPK is now established, including structure-based design, ligand-based design, fragment-based design, as well as structural analysis. This review article highlights the computer-aided drug design approaches utilized to discover of activators targeting AMPK. The principles, advantages or limitation of the different methods are also being discussed together with examples of applications taken from the literatures.