Frontiers in Drug Design and Discovery

The increasing frequency of nosocomial infections due to multiresistant bacterial pathogens represents a serious health concern and is continuously threatening the therapeutic effectiveness of many antibiotics. This medical need calls for the discovery and development of novel antibiotics and for the improvement of existing compounds. Searching for novel chemical classes of antibiotics requires the identification of validated targets for structure-based design or for their transformation into assays suitable for high throughput screening. The power of bacterial genetics and the genomic revolution have provided us with hundreds of targets, which represent components of a bacterial cell essential for viability, well conserved in the desired range of pathogens, and significantly different from mammalian counterparts. In addition, several technological advances in automation and detection systems enable now the transformation of most validated targets into high throughput screening assays. Are these new targets and assays leading to the discovery of promising novel antibiotics? We will review the recent literature for new chemical classes discovered by high throughput screening, describing also the different assays and screening approaches. In addition, we will provide our own considerations on the need to integrate targets and assays with the type and novelty of the chemical diversity, highlighting the power and limitations of high throughput screening for discovering valuable drug leads.

Targeting at the RNA level is considered as an alternative approach to traditional drug discovery focusing on proteins. The targeting of bacterial ribosomal RNA with aminoglycoside antibiotics has provided clear precedence for the targeting of RNA with small molecule drugs. Aminoglycosides can also bind to human cytoplasmic ribosomal RNA and suppress premature termination codons in human messenger RNAs. This suppression activity is explored for the development of novel therapies for genetic disorders caused by premature stop codon mutations, such as cystic fibrosis and Duchenne muscular dystrophy. While aminoglycosides act on ribosomal RNA, certain small molecule metabolites, including vitamins, lysine and purines, can regulate gene expression by binding to messenger RNA at so-called 'riboswitches'. Riboswitches consist of complex three-dimensional structures, located in the 5'-untranslated region of bacterial and fungal messenger RNAs, coding for proteins involved in the uptake, biosynthesis and export of these metabolites. It is unknown whether riboswitches also occur in human cells. Riboswitches provide yet another proof that small molecules can influence biological events by binding directly to RNA. The targeting of human messenger RNAs with small synthetic chemical molecules is a relatively new approach, but may create new and unique opportunities for drug discovery. By using messenger RNA as a target, all genes in the human genome may be considered, including those that encode proteins that are not amenable to highthroughput screening, but for which clear associations with disease processes follow from (molecular) genetic or pharmacological investigations. In this area, progress has been made recently in the targeting of the Alzheimer's b- amyloid precursor protein messenger RNA, and the tumor necrosis factor-a messenger RNA.

The advent in the mid to late 80s of the concept and methods of construction and high throughput screening of phage-displayed combinatorial libraries of proteins and peptides is a revolutionary step in the modern history of molecular biology and its area dedicated to make deep insights into the principles of inter-molecular recognition, the cornerstone of all mechanisms and processes in the living systems. High throughput screening of combinatorial libraries is a miniature model of selection of “best among many candidates” occurring everyday in the nature and resulting in this laboratory scale in new substances, peptides and proteins, with capacities of recognition of their partner molecules of extreme physiological significance required for the basic studies and resolution of chronically persisting and newly emerging biomedical concerns. In this review I intend to give a summary of the emergence of the methodology and its main achievements, focusing on the perspectives it shows for the drug design and discovery via the use of its highly effective protein-peptide engineering and selection techniques. The analysis is directed towards the most troubling infectious and autoimmune diseases (AIDS, hepatitis, cancer, diabetes, parasitological, toxicological). The objective of this review is to give a sketch of the frontiers in drug design and discovery based on the actual state of the research, the tendencies that are seen and the limitations that should be overcome to direct the potential of highthroughput screening technology to resolution of major medical problems.

High throughput screening (HTS) was developed over the last decades with an aim at selecting interesting drug candidate leads within the huge libraries of compounds obtained by combinatorial chemistry. From the early days, the technique consisted in detecting molecules capable of interacting with specifically selected target receptors, enzymes or antibodies so as to sort out the ones susceptible of further development. In the early days, radioactive, enzyme or fluorescent detection techniques were commonly applied in plate reader formats. Throughput was progressively enhanced by increasing the number of wells per plate and by automating the procedures. A further development of the throughput capacity was obtained with the switch from inhomogeneous to homogeneous technologies, where the detection signal is generated directly from the biomolecular interaction, without the need to separate bound from free fractions within the reaction mixture. This advance reduced the manipulation steps, reducing the costs and further improving screening throughput which culminates today in ultra-high-throughput methodologies. The increase in data generated by such technological improvements required a parallel development of softwares capable of handling and analysing the information, at the expense sometimes of missing false negatives. The drug discovery process now faces a new bottleneck resulting directly from HTS developments. Although HTS allows to rapidly identify new drug candidates, the technique does not address the possible applicability of the hits to a biological system. The technology does not give any insight into the possible cellular toxicity of the hit, which is a minimal requirement for further development. Thus, the bottleneck is currently shifting from hit identification to toxicity and biodisponibility evaluation. This new challenge will require a further effort in inventiveness from the part of pharmaceutical scientists.

The hybridization of oligo-DNAs complementary to the sequences of the genes for verotoxins (Shiga toxins) type 1 and 2 of enterohemorrhagic Escherichia coli (EHEC) was monitored using fluorescence polarization under the reaction condition of high salt concentration (0.8 M NaCl), which had been optimized to obtain a high rate of hybridization. The time courses of fluorescence polarization for the fluorescently labeled oligomers (probe DNAs) mixed with the amplified DNA of the genes were recorded. The secondary structures of the amplified single-stranded DNAs were forecasted based on the calculation for minimum free energy at a specified minimum stacking length. Five probe DNA sequences were designed, some of which hybridized extremely rapidly with an amplified product for the gene of Shiga toxin type 1. In the cases using the two different probe DNAs, the hybridization was 90% complete in less than 1 min, considerably faster than that of the 3 min reported previously, while with another probe it was not complete in more than 14 min. The variety of the rate for hybridization could not be explained by melting temperature or G + C content of the probe sequences. It was suggested that the reason for the slow hybridization would be a steric hindrance, comparing the hybridization rates with the shapes around the binding sites for the probes in the secondary structures.

Technologies used for high throughput screening (HTS) at G protein coupled receptors (GPCRs) comprise two major approaches; those generally conducted measuring signal intensity changes using a microtiter plate format, and those measuring cellular protein redistribution via imaging-based analysis systems. Several homogeneous assays, i.e. those without wash and fluid phase separation steps, measure changes of second messenger signaling molecules including cAMP, Ins P3 and calcium. Imaging based assays determined the translocation of GPCR associated proteins such as b arrestin, or internalization of the receptor labeled with fusion tags. Generally, the former assays are used in a primary screening campaign, whilst the latter are used in secondary screening and lead optimization. However, increasing use of automated confocal imaging systems and prevalence of modified cell lines has expanded use of protein redistribution assays. Finally, radiometric techniques are widely used, frequently to measure GPCR ligand binding, using a scintillation proximity assay format. In this paper, the various assay methods used for HTS at G protein coupled receptors are compared and contrasted.

Overviews of the principal techniques that have found application in the profiling of natural products are given, including advances in capillary and column trapping LC-NMR-MS. Single hyphenated spectroscopic techniques, like LC-NMR and LC-MS, currently offer robust and efficient approaches for the rapid dereplication of natural product extracts. Multiple hyphenation techniques, such as (HP)LC-UV-NMR-MS-FTIR, now give an effective and comprehensive method for the deconvolution of complex mixtures. Ongoing improvements include miniaturization and cryogenic NMR probes and the hyphenation to capillary scale LC separations to analyse smaller quantities of samples. Extensions in hyphenation techniques required for natural product and other drug discovery programs include the need for LC-13C NMR and the combination of bioassays with already well-established hyphenated separationspectroscopic techniques, coupled with automated database searching capabilities (data libraries for LC, UV, NMR, MS and other search criteria).

A systematic study of the incidence of kinetics in high throughput screening for drug design and discovery is presented and discussed. This study includes dynamic aspects of the techniques and assays used in this complex research field, kinetic methods applied to the selection of drug candidates, and recent advances in the elucidation of drug-receptor interaction mechanisms and in high throughput pharmacokinetic screening.

The assessment of iDEA pkEXPRESS™ 1.1 for the in silico prediction of Caco-2 permeabilities was investigated. We found that when the software was used in a high/low classification scheme with a cutoff Caco-2 permeability value 5 nm/s, the prediction ability of the tool to evaluate large data sets was acceptable. Using a Caco-2 library of 666 compounds, approximately 78% of the predictions in a high/low classification scheme were correct. In addition, the average fold error of these correct predictions was approximately 3. After removing compounds contained in the training set (275 compounds), the prediction ability of the tool was still acceptable where overall predictions in a high/low classification scheme was 76% correct with an average fold error of approximately 3. The software could be useful for analyzing hits after high-throughput screening of large structurally diverse compound libraries. Unfortunately, the software was unable to provide a measure of confidence to the predictions. Without being able to identify poor predictions, it is difficult to evaluate small data sets using the software.

This study investigated the applicability of metabonomic urinalysis as a toxicological screen in a drug discovery setting as a means to reduce the attrition of new chemical entities in pre-clinical and clinical development. The model hepatotoxins, a-naphthylisothiocyanate (ANIT) and thioacetamide (TA) were dosed in rats to validate our methodologies. The toxins generated significant deviations from the normal metabolic profile of the rat urine, and principal component analysis (PCA) readily showed a separation of controls from animals given acute doses. Spectral changes observed upon ANIT dosing reflected previous literature results showing a large decrease in tricarboxylic acid cycle intermediates and the appearance of bile in the urine. The spectral changes observed for TA reflected glycosuria consistent with known toxicological observations. A test set of drug candidates that failed in preclinical testing, because of target organ toxicity, was then submitted to metabonomic urinalysis. Metabonomic urinalysis failed to classify 3 of the 5 compounds as having toxicological problems. This deficiency can be attributed to the acute dosing protocol and possibly the need to fine tune data collection times. Metabonomic urinalysis as a toxicological screening tool in our environment, a pharmaceutical drug discovery department, appears not to be useful due to a high rate of false positive results and the large amount of material required for a single acute dose in rats.

The main contribution to the water-accessible surface area of lysozyme helices is the hydrophobic term, while the hydrophilic part dominates in the sheet, what is related to the 1-octanol-, cyclohexane- and chloroform-water partition coefficients Po-ch-cf of helices, which are greater than those of the sheet are. The analysis of atom-group partial contributions to logPo-ch-cf allows building local maps. The molecular lipophilicity pattern differentiates among helices, sheet and binding site. For a given atom, logP is sensitive to the presence of other atoms. The contributions of C70-a-c atoms to logP are slightly greater than that of d-e are, which correlate with the distances from the nearest pentagon. (10,10) is the favourite single-wall carbon nanotube (SWNT), presenting consistency between a relatively small aqueous solubility and great Po-ch-cf. Efforts to use fullerenes-SWNTs in therapeutic applications are re-evaluated. There is a strong possibility for hydrophobic interactions between proteins and fullerenes-SWNTs in biomilieu, when the latter is used for biomedical applications, unless the molecule is delivered effectively at the intended site of action.

The past decade has seen a rapid growth in biomedical data from many fields: genetics, chemistry, pharmacology and medicine among others. Structured data integration within and among these fields exists to varying degrees, but unstructured data integration has existed since the dawning of science in the form of text-based published reports. The biomedical literature is vast, with Medline having approximately 15 million records and adding new records at a rate greater than one per minute. Medline contains a wealth of information about chemical compounds, interactions, side-effects, phenotypes, genetic interactions and disease studies. Computational methods are being designed to data mine these large bodies of unstructured text to infer what is not yet known based upon. Applied to drug discovery, this approach has become a potential means of shortcutting the traditional drug discovery “pipeline”, which has been estimated to take up to 15 years and cost approximately 1 billion US$ from target selection to FDA approval. These literature-based methods of knowledge discovery provide a means to identify candidate compounds to treat diseases and to identify genes that may play a role in rare, but extremely adverse reactions to promising new drugs that subsequently force their removal from the pipeline. This chapter discusses the use of literature-based sources of knowledge as a means of discovering novel connections between pharmacological entities such as diseases, drugs and genes.

The role of protein crystallography in drug discovery has changed dramatically during the last 20 years. Based on advances in molecular biology, X-ray techniques, and computation methods, novel structures can be solved in dramatically shorter timescales than were possible previously. These advances have led to the use of structural biology in the earliest stages of drug discovery, lead identification and optimization. Advances have also decreased the time needed to solve protein-ligand structures from weeks to hours. In the past, typically only a handful of co-crystal structures for an individual project were determined during the discovery phase, and frequently, these structures were available too late to greatly impact chemistry efforts. Today, hundreds of compounds can be screened crystallographically to determine if they bind. The reduction in turnaround time for crystal structures has led to a more significant contribution of Structure-Based Drug Design during the earliest phases of drug discovery. This review will examine the ways in which the new technologies, fragment-based design, structure-based combinatorial chemistry, and the structure of macrocyclic inhibitor complexes have impacted early phase drug discovery.

Whole gene synthesis is rapidly becoming a powerful technology that allows researchers the ability to distill a growing body of genetic and structural information into improved nucleic acid sequences that would otherwise be impossible to obtain by traditional cloning and mutagenesis methods. Recent advances in the efficient small-scale manufacture of long and accurate oligodeoxyribonucleotides has resulted in a low cost source of building blocks for the assembly of larger DNA molecules by polymerase chain reaction methods. Proof of concept experiments have yielded synthetic replication competent viral genomes, as well as synthetic multi-gene clusters of greater than 30 kilobase pairs in size encoding multi-enzyme systems that catalyze efficient biosynthesis of small drug molecules. These advances have placed whole gene synthesis on a cost trajectory that will lead to unprecedented advances in synthetic biology, ranging from the engineering of protein crystals to the production of re-engineered translation machineries that can produce totally novel protein-like materials. The possible advances in synthetic biology, enabled by whole gene synthesis, will be limited only by the imagination of the applied life sciences research community.