Current Pharmaceutical Design - Volume 13, Issue 34, 2007

Discovery of effective and safe drugs traditionally entails the synthesis and/or screening of large numbers of compounds. Considering that marketed compounds represent only one compound out of every 30,000 compounds synthesized, the pharmaceutical industry has devoted substantial efforts and finances to methodologies that have the potential to shorten the discovery process. One of the best strategies is to use “Rational Drug Design” to shorten the discovery time by dramatically narrowing the pool of potential drug candidates through calculation of drug-protein binding interaction energies. Although highly appealing in theory, in practice the efforts devoted to rational drug design in the 1980s and 1990s resulted in predictions that were frequently inaccurate due to the approximations and assumptions employed in the calculations. Advances in X-ray crystallography, NMR and extraordinary advances in available CPU computer power over the past decade have resulted in more accurate predictions. Greater accuracy is achieved through the use of higher levels of quantum mechanical theory to provide a more extensive and accurate set of molecular mechanics force field parameters for protein residues as well as small molecules, by inclusion of solvent effects, and by using computer simulation methods that exhaustively search conformational space. Improvements in computer simulation methods such as Molecular Dynamics and Monte Carlo and available 3-dimensional structures of protein-ligand complexes enable calculation of free energy differences which are important, since free energy differences are directly related to the experimental result. This issue on “Rational Drug Design” covers the recent advances in Computer Aided Drug Design (CADD) Methods as well as their successful application to a variety of drug discovery programs. The first article [1] focuses on various computational aspects for identification of leads to drug targets in silico. In addition the article discusses the fundamental issues and challenges associated with various CADD methods. The second article [2] describes the dynamical behavior of the binding pocket S1 in the apo forms of metalloproteinase types 2 and 3 using molecular dynamics simulations. Results from this study are useful in the design of specific metalloproteinase inhibitors. The third article [3] summarizes the role of unconventional hydrogen bonds in the recognition of small molecules by biological receptors of pharmaceutical relevance. The fourth article [4] describes predictive QSAR modeling and virtual screening of small molecule databases for several drug targets. The fifth article [5] focuses on lead inhibitor optimization strategies using the free energy perturbation approach and molecular mechanics methods and evaluates the merits of each method for predicting relative binding affinities of COX-2 inhibitors. The sixth article [6] summarizes use of molecular modeling and informatics tools for the discovery of anti-diabetic agents. The last article [7] reviews rational drug design strategies for development of antiviral agents directed against the influenza virus replication. Overall the issue provides computational and medicinal chemists in both academia and industry an extensive overview of the scope and limitations of CADD methods useful for rational drug design. As an Executive Editor of Current Pharmaceutical Design, I would like to thank all the authors for contributing to this issue on Rational Drug Design. I would also like to thank Dr. Mark Erion for his helpful suggestions, encouragement and support in editing this issue. References [1] Shaikh SA, Jain T, Sandhu G, Latha N, Jayaram B. From drug target to leads-sketching a physicochemical pathway for lead molecule design in silico. Curr Pham Des 2007; 13(34): 3454-3470. [2] de Oliveira CAF, Zissen M, Mongon J, Mccammon JA. Molecular dynamics simulations of metalloproteinases types 2 and 3 reveal differences in the dynamic behavior of the S1' binding pocket. Curr Pham Des 2007; 13(34): 3471-3475. [3] Toth G, Bowers SG, Truong AP, Probst G. The role and significance of unconventional hydrogen bonds in small molecule recognition by biological receptors of pharmaceutical relevance. Curr Pham Des 2007; 13(34): 3476-3493..

The discovery of new pharmaceuticals via computer modeling is one of the key challenges in modern medicine. The advent of global networks of genomic, proteomic and metabolomic endeavors is ushering in an increasing number of novel and clinically important targets for screening. Computational methods are anticipated to play a pivotal role in exploiting the structural and functional information to understand specific molecular recognition events of the target macromolecule with candidate hits leading ultimately to the design of improved leads for the target. In this review, we sketch a system independent, comprehensive physicochemical pathway for lead molecule design focusing on the emerging in silico trends and techniques. We survey strategies for the generation of candidate molecules, docking them with the target and ranking them based on binding affinities. We present a molecular level treatment for distinguishing affinity from specificity of a ligand for a given target. We also discuss the significant aspects of drug absorption, distribution, metabolism, excretion and toxicity (ADMET) and highlight improved protocols required for higher quality and throughput of in silico methods employed at early stages of discovery. We present a realization of the various stages in the pathway proposed with select examples from the literature and from our own research to demonstrate the way in which an iterative process of computer design and validation can aid in developing potent leads. The review thus summarizes recent advances and presents a viewpoint on improvements envisioned in the years to come for automated computer aided lead molecule discovery.

Matrix Metalloproteinases (MMPs) are zinc-containing proteinases that are responsible for the metabolism of extracellular matrix proteins. Overexpression of MMPs has been associated with a wide range of pathological diseases such as arthritis, cancer, multiple sclerosis and Alzheimer's disease. The excessive and unregulated activity of Matrix Metalloproteinases type 2 (MMP-2), also known as gelatinase A, has been identified in a numbers of cancer metastases. Several MMP inhibitors (MMPi) have been proposed in the literature aiming to interfere in the MMPs activity. In this work we performed long MD simulations in order to study the dynamical behavior of the binding pocket S1' in the apo forms of MMP type 2 and 3, and identify, at the molecular level, the structural properties relevant for the designing of specific inhibitor of MMP-2.

The discovery and optimization of nonbonded interactions, such as van der Waals interactions, hydrogen bonds, salt bridges and the hydrophobic effect, between small molecule ligands and their receptors is one of the main challenges in rational drug discovery. As the theory of molecular interactions advances more evidence accumulates that nonbonded interactions, such as unconventional hydrogen bonds (X-H Y interactions, where X can be either C, N or O atom and Y can be either an aromatic ring system, O or F atom), contribute to ligand recognition by biological receptors. This review provides an overview of unconventional hydrogen bonds between ligands and their receptors of pharmaceutical relevance by dissecting their structure activity relationships and 3D structural elements. Gaining an understanding of the energetic and the structural properties of unconventional hydrogen bonds in ligand-receptor interactions leads us to the elucidation of their practical significance. Ultimately, this enables us to consciously apply these interactions in hit and lead optimization in rational structure based drug design.

Quantitative Structure Activity Relationship (QSAR) modeling has been traditionally applied as an evaluative approach, i.e., with the focus on developing retrospective and explanatory models of existing data. Model extrapolation was considered if only in hypothetical sense in terms of potential modifications of known biologically active chemicals that could improve compounds' activity. This critical review re-examines the strategy and the output of the modern QSAR modeling approaches. We provide examples and arguments suggesting that current methodologies may afford robust and validated models capable of accurate prediction of compound properties for molecules not included in the training sets. We discuss a data-analytical modeling workflow developed in our laboratory that incorporates modules for combinatorial QSAR model development (i.e., using all possible binary combinations of available descriptor sets and statistical data modeling techniques), rigorous model validation, and virtual screening of available chemical databases to identify novel biologically active compounds. Our approach places particular emphasis on model validation as well as the need to define model applicability domains in the chemistry space. We present examples of studies where the application of rigorously validated QSAR models to virtual screening identified computational hits that were confirmed by subsequent experimental investigations. The emerging focus of QSAR modeling on target property forecasting brings it forward as predictive, as opposed to evaluative, modeling approach.

Cyclooxygenases (COXs), the enzymes involved in the formation of prostaglandins from polyunsaturated fatty acids such as arachidonic acid, exist in two forms-the constitutive COX-1 that is cytoprotective and responsible for the production of prostaglandins and COX-2 which is induced by cytokines, mitogens and endotoxins in inflammatory cells and responsible for the increased levels of prostaglandins during inflammation. As a result COX-2 has become the natural target for the development of anti-inflammatory and anticancer drugs. While the conventional NSAIDs with gastric side effects inhibit both COX-1 and COX-2, the newly developed drugs for inflammation with no gastric side effects selectively block the COX-2 enzyme. NSAIDs, nonselective non-aspirin NSAIDs and COX-2 selective inhibitors, are being widely used for various arthritis and pain syndromes. Selective inhibitors of COX-2, however, convey a small but definite risk of myocardial infarction and stroke; the extent of which varies depending on the COX-2 specificity. In view of the gastric side effects of conventional NSAIDs and the recent market withdrawal of rofecoxib and valdecoxib due to their adverse cardiovascular side effects there is need to develop alternative anti-inflammatory agents with reduced gastric and cardiovascular problems. The present study reviews various Computer Aided Drug Design (CADD) approaches to develop Cyclooxygenase based anti-inflammatory and anti-cancer drugs.

Diabetes mellitus is a chronic metabolic disorder, characterized by glucose overproduction and glucose underutilization. Current therapy for T2DM includes drugs, like metformin, glitazones, sulphonyl ureas, etc. Extensive research has been carried out world wide on molecular targets for T2DM like PPARγ, PTP1B, DPP-IV, GSK-3, cannabinoid receptor, fructose-bisphosphatases, β3 adrenoceptor, etc. in the development of newer anti-diabetic agents. These therapeutic targets are quite important and most of them are suitable for in silico analysis. Hence, many molecular modeling and informatics studies like, molecular docking, pharmacophore mapping, 3DQSAR, virtual screening, quantum chemical studies, and pharmacoinformatics like bioinformatics and chemoinformatics studies have been performed on the drugs / leads / targets associated with T2DM. Several of these in silico efforts are exemplary studies; the methodologies adopted in these studies can be emulated in many other therapeutic areas. A review of the rational approaches reported in designing anti-diabetic agents is presented in this article.

In this review, we will discuss drug design based on proven and potential anti-influenza drug targets including viral hemagglutinin (HA), neuraminidase (NA), M2 ion channel, 3P polymerase complex, and host factors such as kinases. We have summarized influenza inhibitors based on their mode of actions. For instance, included are descriptions of (1) inhibitors of HA cleavage, such as nafamostat, camostat, gabexate, epsilon-aminocapronic acid and aprotinin, (2) inhibitors of fusion and entry, such as benzoquinones and hydroquinones, CL 385319, BMY-27709, stachyflin, and their analogues, (3) inhibitors of viral RNPs/polymerase/endonuclease, such as T-705, L-735,822, flutimide and their analogues, (4) inhibitors of MEK, such as PD 0325901, CI-1040 and ARRY-142886, and (5) inhibitors of NA such as DANA, FANA, zanamivir, and oseltamivir, etc. Although amantadine and rimantadine are not recommended for treating influenza virus infections because of drug resistance problem, these viral M2 ion channel blockers established a proof-of-concept that the endocytosis of virion into host cells can be a valid drug target because M2 protein is involved in the endocytosis process. The influenza polymerase complex not only catalyzes RNA polymerization but also encodes the “cap snatching” activity. After being exported from the nucleus to the cytoplasm, the newly synthesized vRNPs are assembled into virions at the plasma membrane. The progeny virions will then leave the host cells through the action of NA. The strategies for discovery of small molecule inhibitors of influenza virus replication based on each particular mechanism will be discussed. Finally, the lessons learned from the design of NA inhibitors (NAI) are also included. Many exciting opportunities await the cadre of virologists, medicinal chemists, and pharmacologists to design novel influenza drugs with favorable pharmacological and pharmacokinetic properties to combat this threatening infectious disease.