Combinatorial Chemistry & High Throughput Screening - Volume 11, Issue 7, 2008

Although in vitro biochemical assays, such as enzyme activity and receptor binding, have been used extensively in the past to discover new drugs, there is a rapid increase in the use of assays based on living cells. Cell-based assays provide a target in a more physiologically relevant environment than biochemical assays. According to a worldwide survey involving more than 50 pharmaceutical and biotech companies, more than 50% of all primary screens are currently cell-based [1]. Due to the use of cryopreserved and division-arrested cells, cell-based screening also has become more efficient and more flexible [2]. This special issue of Combinatorial Chemistry & High Throughput Screening is devoted to cell-based screening. Review articles were collected on screening in live cells of all major drug target classes, i.e. receptors, ion channels and protein kinases. New technology developments are discussed, such as high content analysis, label-free assays and the use primary and embryonic stem cells. In the first article, Jorg Hüser and colleagues from Bayer HealthCare AG (Wuppertal, Germany) review the technological approaches used for ultra-High Throughput Screening (ultra-HTS). ‘Ultra’ means automated screening of more than 100,000 data points per day. To differentiate target hits from non-specifically acting compounds, integral reference signals are incorporated into the ultra-HTS assays. George Hanson and Bonnie Hanson from Invitrogen Discovery Sciences (Madison, WI) give an overview of fluorescent probes and types of fluorescent assays that are applied in cellular assays for a number of pharmaceutically relevant target classes, including protein kinases and ion channels. They also discuss cellular pathway analysis and the combining of multiple read-outs in one assay (multiplexing). With recent technological advances in fluorescent probes, the search for novel therapeutics targeting ion channels is accelerating. John Dunlop and colleagues of Wyeth Research (Princeton, NJ) review the various techniques used in screening ion channel targets. Real-life examples of the screening of a ligand-gated ion channel and a voltage-gated channel are presented to demonstrate the utility of fluorescence-based screening. ‘High content analysis’ refers to techniques involving the multiplexed analysis of fluorescent markers to measure multiple cellular responses to biological stimuli or drug treatment at the single-cell level. High content analysis is usually based on automated microscopy and provides multiparametric information on single cells within a population. Fabio Gasparri and colleagues from Nerviano Medical Sciences S.r.l. (Nerviano, Italy) review the concepts and techniques of high content analysis of protein kinases, with an emphasis on kinases implicated in oncology. The assay methods are illustrated with data from the author's laboratory of research on cell cycle kinase inhibitors. Morten Præstegaard and colleagues of Thermo Fisher Scientific Inc. (Soborg, Denmark) describe different strategies of multiplexing green fluorescent protein-based and immunofluorescence translocation assays. The authors differentiate between multiplexing of readouts in the same signal transduction pathway (vertical multiplexing) and of readouts across different signal transduction pathways (horizontal multiplexing). Examples are shown of multiplexing assays in the p38 MAPK pathway and of the activation and internalization of G protein-coupled receptors. Jim Inglese and colleagues of the Chemical Genomics Center of the National Institutes of Health (Bethesda, MD) present a case study of the screening of the glucocorticoid receptor with a translocation assay based on enzyme fragment complementation. A ‘quantitative’ screening approach was used, which means that compounds were screened at multiple concentrations in the primary run. Quantitative HTS increases the information content of HTS. Richard Eglen of PerkinElmer (Waltham, MA), Annette Gilchrist of Caden Biosciences (Madison, WI) and Terry Reisine review the disparities that have been found in the pharmacology of compounds acting on G protein-coupled receptors in recombinant cells used for screening and in natural tissues. The efficiency of identifying effective therapeutics may increase when natural tissues are used more in the drug discovery process. In the following article, Eglen et al. review the use of primary and embryonic stem cells for screening. Human stem cells offer unique opportunities in that they can be directed to specific phenotypes, providing a framework to identify tissue-selective agents. In the last article, Lisa Minor from Johnson & Johnson Pharmaceutical Research Institute (Spring House, PA) reviews the application of electrical impedance and refractive index to measure cell-based functional response. These label-free technologies are rapidly generating interest because they allow measuring phenotypic responses without the addition of exogenous labels or extraction of the cells.

Functional cell-based assays have gained increasing importance for microplate-based high throughput screening (HTS). The use of high-density microplates, most prominently 1536-well plates, and miniaturized assay formats allow screening of comprehensive compound collections with more than 1 million compounds at ultra-high throughput, i.e. in excess of 100,000 samples per day. uHTS operations with numerous campaigns per year should generally support this throughput at all different steps of the process, including the underlying compound logistics, the (automated) testing of the corporate compound collection in the bioassay, and the subsequent follow-up studies for hit confirmation and characterization. A growing number of reports document the general feasibility of cell-based uHTS in microliter volumes. In addition, full automation with integrated robotic systems allows the realization of also complex assay protocols with multiple liquid handling and signal detection steps. For this review, cell-based assays are categorized based on the kinetics of the cellular response to be quantified in the test and the readout method employed. Thus, assays measuring fast cellular responses with high temporal resolution, e.g., receptor mediated calcium signals or changes in membrane potential, are at one end of this spectrum, while tests quantifying cellular transcriptional responses mark the opposite end. Trends for cell-based uHTS assays developed at Bayer-Schering Pharma are, first, to incorporate assay integral reference signals allowing the experimental differentiation of target hits from non-specifically acting compounds, and second, to make use of kinetic, real-time readouts providing additional information on the mode-of-action of test compounds.

A fluorescent probe is a fluorophore designed to localize within a specific region of a biological specimen or to respond to a specific stimulus. Fluorescent probes have been used for nearly a century to study cellular processes due to their exquisite sensitivity and selectivity. Fluorescent probes have also gained in popularity as safety and environmental concerns over the use of radioactive probes have grown. At the same time, cellular assays are being more widely used now than ever before. This review will give a broad overview of types of fluorescent probes, types of fluorescent assays, and their application in cellular assays for a number of pharmaceutically relevant target classes.

Ion channels are attractive targets for drug discovery with recent estimates indicating that voltage and ligandgated channels account for the third and fourth largest gene families represented in company portfolios after the G protein coupled and nuclear hormone receptor families. A historical limitation on ion channel targeted drug discovery in the form of the extremely low throughput nature of the gold standard assay for assessing functional activity, patch clamp electrophysiology in mammalian cells, has been overcome by the implementation of multi-well plate format cell-based screening strategies for ion channels. These have taken advantage of various approaches to monitor ion flux or membrane potential using radioactive, non-radioactive, spectroscopic and fluorescence measurements and have significantly impacted both high-throughput screening and lead optimization efforts. In addition, major advances have been made in the development of automated electrophysiological platforms to increase capacity for cell-based screening using formats aimed at recapitulating the gold standard assay. This review addresses the options available for cell-based screening of ion channels with examples of their utility and presents case studies on the successful implementation of high-throughput screening campaigns for a ligand-gated ion channel using a fluorescent calcium indicator, and a voltage-gated ion channel using a fluorescent membrane potential sensitive dye.

High-content analysis (HCA) is a term used to describe techniques involving multiplexed analysis of fluorescent markers to measure multiple cellular responses to biological stimuli or drug treatment. HCA is usually based on automated microscopy or related technologies, and its value lies in providing multiparametric information on single cells within a population. During the last decade, several HCA approaches have been developed and applied to assess cellular mechanism of action of pharmacologically relevant compounds identified through biochemical screening or similar in vitro methods. With automation and instrument development, these approaches have evolved to the extent that the technique is now routinely used in screening applications, including primary HTS on compound collections. Here, we review the field and discuss in particular the application of HCA to the discovery of small molecule inhibitors targeting kinases which are implicated in Oncology.

Multiplexing of GFP based and immunofluorescence translocation assays enables easy acquisition of multiple readouts from the same cell in a single assay run. Immunofluorescence assays monitor translocation, phosphorylation, and up/down regulation of endogenous proteins. GFP-based assays monitor translocation of stably expressed GFP-fusion proteins. Such assays may be multiplexed along (vertical), across (horizontal), and between (branch) signal pathways. Examples of these strategies are presented: 1) The MK2-GFP assay monitors translocation of MK2-GFP from the nucleus to the cytoplasm in response to stimulation of the p38 pathway. By applying different immunofluorescent assays to the MK2 assay, a multiplexed HCA system is created for deconvolution of p38 pathway activation including assay readouts for MK2, p38, NFκB, and c-Jun. 2) A method for evaluating GPCR activation and internalization in a single assay run has been established by multiplexing GFP-based internalization assays with immunofluorescence assays for downstream transducers of GPCR activity: pCREB (cAMP sensor), NFATc1 (Ca2+ sensor), and ERK (G-protein activation). Activation of the AT1 receptor is given as an example. 3) Cell toxicity readouts can be linked to primary readouts of interest via acquisition of secondary parameters describing cellular morphology. This approach is used to flag cytotoxic compounds and deselect false positives. The ATF6 Redistribution assay is provided as an example. These multiplex strategies provide a unique opportunity to enhance HCA data quality and save time during drug discovery. From a single assay run, several assay readouts are obtained that help the user to deconvolute the mode of action of test compounds.

Nuclear translocation is an important step in glucocorticoid receptor (GR) signaling and assays that measure this process allow the identification of nuclear receptor ligands independent of subsequent functional effects. To facilitate the identification of GR-translocation agonists, an enzyme fragment complementation (EFC) cell-based assay was scaled to a 1536-well plate format to evaluate 9,920 compounds using a quantitative high throughput screening (qHTS) strategy where compounds are assayed at multiple concentrations. In contrast to conventional assays of nuclear translocation the qHTS assay described here was enabled on a standard luminescence microplate reader precluding the requirement for imaging methods. The assay uses β-galactosidase α complementation to indirectly detect GR-translocation in CHO-K1 cells. 1536-well assay miniaturization included the elimination of a media aspiration step, and the optimized assay displayed a Z' of 0.55. qHTS yielded EC50 values for all 9,920 compounds and allowed us to retrospectively examine the dataset as a single concentration-based screen to estimate the number of false positives and negatives at typical activity thresholds. For example, at a 9 μM screening concentration, the assay showed an accuracy that is comparable to typical cell-based assays as judged by the occurrence of false positives that we determined to be 1.3% or 0.3%, for a 3σ or 6σ threshold, respectively. This corresponds to a confirmation rate of ∼30% or ∼50%, respectively. The assay was consistent with glucocorticoid pharmacology as scaffolds with close similarity to dexamethasone were identified as active, while, for example, steroids that act as ligands to other nuclear receptors such as the estrogen receptor were found to be inactive.

For most membrane-bound molecular targets, including G protein linked receptors (GPCRs), the optimal approach in drug discovery involves the use of cell based high throughput screening (HTS) technologies to identify compounds that modulate target activity. Most GPCRs have been cloned and can therefore be routinely expressed in immortalized cell lines. These cells can be easily and rapidly grown in unlimited quantities making them ideal for use in current HTS technologies. A significant advantage of this approach is that immortalized recombinant cells provide a homogenous background for expression of the target which greatly facilitates consistency in screening, thus allowing for a better understanding of the mechanism of action of the interacting compound or drug. Nonetheless, it is now evident that numerous disparities exist between the physiological environment of screening systems using recombinant cells and natural tissues. This has lead to a problem in the validity of the pharmacological data obtained using immortalized cells in as much as such cells do not always reflect the desired clinical efficacy and safety of the compounds under examination. This brief review discusses these issues and describes how they influence the discovery of drugs using modern HTS.

Cellular technologies are widely used in drug discovery to treat human diseases. Most studies involve the expression of recombinant targets in immortalized cells and measure drug interactions using simple, quantifiable responses. Such cells are also amenable to high throughput screening (HTS) methods. However, the cell phenotype employed in HTS is often determined by the assay technology available, rather than the physiological relevance of the cell background. They are, therefore, suboptimal surrogates for cells that accurately reflect human diseases. Consequently, there is growing interest in adopting primary and embryonic stem cells in drug discovery. Primary cells are already used in secondary screening assays in conjunction with confocal imaging techniques, as well as in target validation studies employing, for example, gene silencing approaches. Stem cells can be grown in unlimited quantities and can be derived from transgenic animals engineered to express disease causing proteins better coupling the molecular target with function in vivo. Human stem cells also offer unique opportunities for drug discovery in that they can be directed to specific phenotypes thus providing a framework to identify tissueselective agents. Organizing stem cells into networks resembling those in native tissues, potentially returns drug discovery back to the highly successful pharmacological methods of the past, in which organ and tissue based systems were used, but with the advantage that they can be utilized using modern HTS technologies. This emerging area will lead to discovery of compounds whose effect in vivo is more predictable thereby increasing the efficiency of drugs that ameliorate human disease.

Label-free technologies based on electrical impedance or refractive index are new tools for measuring a cellbased functional response. Although the technologies are relatively new to high throughput screening cell-based applications, they are rapidly generating interest in that they are able to measure a phenotypic response using cells natively expressing the target protein without using dyes or cellular extracts. In addition, one can measure the cellular response using a kinetic mode resulting in an assay potentially rich in content. This article will describe these technologies and their applications in measuring cell proliferation, cell attachment and spreading, cell apoptosis and their application for several receptor target classes, including receptor tyrosine kinases and G protein-coupled receptors. The potential utility and drawbacks of these tools for high throughput screening, directed screening and compound profiling will also be discussed.

Full text available