A powerful tool for drug discovery

Receptor-ligand binding assays are extremely powerful tools in drug discovery. With advances in technology and methodology, the traditional radioligand filtration assays are being replaced with higher throughput homogeneous assays. Non-radioactive methods have been developed recently that are particularly applicable to early phase drug discovery.

The physiological and pathological effects of neurotransmitters and hormones are mediated by specific receptors. Thus, modulating the actions of receptors such as G protein-coupled receptors (GPCR), ligand-gated ion channel receptors, tyrosine kinase receptors and nuclear hormone receptors (NHR) have been a central focus in the quest to intervene pharmacologically in the treatment of disease. Receptor-ligand binding assays provide one of the most powerful techniques with which to aid structure-activity relationship studies to improve potency and efficacy, to assist in validating rational drug design and to elucidate the mechanism of action and selectivity of known drugs. Such studies help improve upon first generation medicines and produce novel drugs to enhance human health.

Classical pharmacological studies monitored the properties of receptor-ligand interaction based on the receptor-mediated response. The advent of radiolabeled ligands during the 1970s opened the door to the direct identification of biological receptors1. Widespread adoption of radioligand binding methods led to an explosive growth in the discovery of receptor types (and subtypes) as well as to new insights into the dynamic regulation and signal transduction of receptors. Today, radioligand binding is still used as a gold standard method in receptor research. The technical ease and throughput of radioligand binding assays have been greatly enhanced by the availability of microplates, harvesters and readers. Combined with automated pipetting workstations, conventional radioligand binding assays are commonly used to screen tens of thousands of compounds in a drug discovery campaign (typically ~2,000 data points/day), although the filtration and washing steps in the assays are still labour-intensive and time-consuming.

With the rapid increase in the size of compound libraries as well as new targets emerging from the Human Genome Project, scientists are demanding assays with higher throughput to test more potential chemical starting points against many disease targets and to do so in a short time period. Secondary screening and Lead Generation, where a large number of ‘hit’ compounds are refined to a viable set of ’lead‘ series, also requires a receptor-binding assay with a rapid turnaround of results. Significant technological advances have already enabled the design of receptor-ligand binding assays for high throughput screening (HTS). For example, conversion of the radioligand binding to a so-called ’homogeneous‘ assay format, which has no separation steps, provides a simple radioligand binding assay suitable for fully automated HTS, to enable processing 20,000-100,000 data points per day. Besides the radioligand assays, non-radioactive approaches such as fluorescent ligand binding and label-free affinity binding assays have attracted great attention in receptor research.

Filtration-based radioligand binding

The classical filtration-based radioligand-binding assay is quite simple in design. A preparation containing the receptor of interest is incubated with a suitable radioligand for an appropriate period of time. The reaction is terminated by filtration to remove the large excess of unbound radioligand; the amount of radioligand bound to the receptor is then quantified. For membrane-bound receptors such as GPCRs, one typically uses membrane preparations for receptor binding, although whole cell binding can be performed to address specific questions that require intact cells. To stop the binding reaction the receptor preparation is collected on filters and rapidly washed to remove unbound radioligand. Today, most binding assays are performed in 96-well microplate filtration formats. Recently, Millipore has developed a 384-well filtration format that increases assay throughput. Practical guidelines of developing radioligand-binding assays for membrane-bound receptors have been reviewed in greater detail recently2.

Filtration is often used to separate unbound from bound radioligand for soluble receptors such as NHR. Washing steps, however, may not only cause dissociation of the receptor-ligand complex, but also generate large volumes of radioactive waste. In an effort to circumvent these liabilities, a centrifugation-based thin layer gel filtration chromatographic (GFC) method has been developed in a 96-well format. This method has been successfully applied to radioligand NHR binding assays3. The method exhibits excellent recovery of protein-ligand complexes and less opportunity for dissociation of the complex since it eliminates major dilution effects from the mobile phase of a column and from elution steps used in conventional GFC.

Homogeneous radioligand binding

One of the most significant advances in radioligand binding techniques was the development of the scintillation proximity assay (SPA)4. The core of the technology is a bead containing a scintillant. When radioligand binds to a receptor captured on the surface of the bead, it becomes close enough to stimulate the scintillant to emit light. Unbound radioligand stays in bulk solution and does not come in close proximity to the scintillant, contributing a low background signal. SPA does not require a physical separation of unbound from bound radioligand, thus reducing the potential for shifting the apparent equilibrium of the binding event under study, as well as potential liquid-handling error. The removal of the separation step also means that SPA is faster and more simple to perform, easy to automate and generates much less radioactive waste.

SPA beads are available with several different coatings to capture target molecules on the bead surface. For membrane-bound receptors, wheat-germ agglutinin (WGA) coated beads are most often used, since WGA binds carbohydrate residues on glycosylated proteins. However, for the same reason, WGA beads should not be used with glycoprotein ligands (e.g. platelet-derived growth factor) because these ligands will bind directly to the bead. Polylysine coated beads are also used to capture negatively charged membranes via their positively charged properties. Similar electrostatic interactions may also occur between polylysine and receptor or ligand, which may affect the receptor-ligand interaction that is being studied. Therefore, it is important to validate the assay with a list of known agonists and antagonists with varying affinities.

Different core materials have been used in manufacturing SPA beads. Yttrium silicate and red-shifted yttrium oxide have the best scintillation efficiency and as such, result in the highest signal output. For example, WGA-coated yttrium beads gave a two fold higher signal-to-background window compared to other bead types when used to quantify adenosine 2a receptor binding5. However, yttrium beads have a density (4.1 g/cm3) that is much higher than that of a conventional aqueous buffer so that the particles settle quickly in solution. Stirring or shaking is needed when pipetting bead suspensions to ensure an even dispensing of the beads and also during the assay incubation period to facilitate obtaining equilibration. However, one must be cautious because vigorous stirring or shaking may cause denaturation of the receptor protein. Adding glycerol to the bead suspension is an alternative that can reduce the rate of bead settling. Although their scintillation efficiency is not quite as high as yttrium beads, polyvinyltoluene beads have a density (1.05 g/cm3) closer to that of aqueous buffer, which means that the beads remain suspended much longer in aqueous solution.

The SPA application is based on achieving a balance between the energy of the decay particles released from the radioligand and the distance these particles must travel to stimulate a scintillant. Although several radioisotopes have been used successfully with SPA, tritium (3H) is ideally suited because its β-particle has an extremely short pathlength through water. This means that the background from unbound 3H is normally low, even when relatively large amounts of radioactivity are used. However, the low energy of the bparticle means lower efficiency of energy transfer in SPA. As a result, some [3H] radioligand binding assays work well in filtration format but do not have sufficient signals in SPA. Energy transfer from Iodine-125 (125I) to beads is not an issue since 125I releases higher energy Auger electrons. But these electrons emitted from unbound isotopes will stimulate beads due to their long pathlengths, resulting in a higher level of background. Packing beads together via settling or centrifugation will increase energy transfer between bound radioligand and scintillant in the beads because some electrons or β-particles can travel beyond their bead of origin and stimulate adjacent beads as well.

SPA is a solid phase assay; therefore, it is important to optimise the ratio between membranes and beads. If a tissue is to be used, it should be a relatively rich source of the receptor of interest due to the limitations of the coupling capacity of the SPA beads. Usually, precoupling membranes to SPA beads will give higher specific binding with lower non-specific binding compared to simultaneous additions of membranes, beads and radioligand.

Because SPA is a homogeneous assay, one can read the samples at increasing time intervals until the specific signal stabilises, indicating equilibrium has been obtained. It is important to establish both the incubation time required for reaching equilibrium and the stability of the equilibrium. This will determine the time window for recording the signal and how many samples can be processed in one experiment.

Colour quenching is a problem in radioligand binding, because compound libraries invariably contain coloured samples. Coloured compounds absorb a proportion of the light emitted from the scintillant, thereby appearing as false positives in the assay. Colour quenching is troublesome in SPA format because the test compound remains present when the sample is read. In principle, this liability can be overcome by using colour quench correction with the appropriate scintillation counter. But in practice, such a correction results in higher error due to the steep slope of the correction curve.

FlashPlate® technology is similar in concept to SPA. The interior of each FlashPlate® well has a thin layer of polystyrene-based scintillant. The FlashPlate® is then coated with WGA or specific antibody to capture soluble receptor or receptor-containing membranes. Only receptor-bound radioligands are in close enough proximity to stimulate the scintillant to emit light, thereby providing another homogeneous platform for HTS. Aspiration and wash steps can be added to the FlashPlate® assay protocol to improve signal-to-background ratios.

Fluorescent ligand binding

Non-radiometric homogeneous receptor binding assays have been developed, in part to reduce the use of radioactive ligands. Fluorescence-based technologies have become increasingly popular for quantifying receptor-ligand interaction because its spectral characteristics are very sensitive to environmental changes. For the same reason, fluorescence methods are much more prone to artifacts in compound screening, e.g. autofluorescence, quenching and inner-filter effects.

Fluorescence polarisation, which is based on changes of molecular rotation rates upon binding of fluorescent ligands to receptors, has been successfully used for HTS of both soluble and membrane-bound receptors6,7.

A fluorescence-based thermal shift assay has been described that uses a fluorescent dye to monitor receptor thermal unfolding8. Ligand binding affinity can be assessed from the shift of the unfolding temperature obtained in the presence of ligand, which stabilises the receptor confirmation, relative to that obtained in the absence of ligand. Microplate thermal shift assay has been developed for HTS of soluble receptors9, but has not yet worked well for membrane-bound receptors.

Capillary electrophoresis (CE) coupled to laser-induced fluorescence detection has been developed to detect receptor-ligand interactions10. Briefly, soluble receptor migrates in the capillary when an electric field is applied. Binding of a ligand to the receptor imparts a change in charge or conformation resulting in a shift in its mobility, which is detected using laser-induced fluorescence. Affinity CE has been developed for HTS11, but only works for soluble receptors.

Recent developments in confocal microscopy have been exploited to study interactions of fluorescently labeled ligands with receptors at single-molecule detection sensitivity. Fluorescence correlation spectroscopy (FCS) and fluorescence intensity distribution analysis (FIDA) are two different approaches that quantify fluctuations in fluorescence intensity within a microenvironment. FCS analyses the diffusion time of fluorescent molecules in a small illuminated sample volume. FIDA quantifies changes in fluorescence intensity induced by binding of fluorescent ligands to receptors in membranes. Because they utilise confocal detection technology, FCS and FIDA inherently allow reduction of the assay volume without any loss of signal. Therefore, they are attractive in miniaturised HTS12-14. The comprehensive theory and practical considerations of these novel fluorescent methods have been described recently15.

Label-free receptor binding

The majority of techniques currently employed to interrogate a biomolecular interaction require some type of labeling, typically a fluorescent molecule, enzyme, or radioisotope, to report the binding event. However, there are some techniques that do not require labeling of the ligand or the receptor and thereby circumvent the concern that the label itself perturbs the receptor-ligand interaction.

Surface plasmon resonance (SPR) is a label-free detection technology to directly measure receptor-ligand interactions. In brief, SPR is based upon light refracting off the surface of a gold plated sensor chip. The refraction angle of the light is sensitive to increases in mass at the surface of the chip16. For example, SPR has been used to measure the change in refractive index as NHR binds to a ligand covalently linked to the chip17. In screening mode, compounds bind to the receptor and thereby prevent the receptor-ligand interaction on the chip. This technology has been successfully utilised to detect very weak receptor binders (Kd ~ 1 mM).

When coupled to a separation technique, mass spectrometry (MS) can be used to identify compounds that bind to a soluble receptor. For example, GFC is used to separate bound compounds from unbound. The receptor-ligand complexes are subsequently trapped on a reversed phase liquid chromatography (LC), after which dissociation and concentration takes place. The trapped molecules are then eluted and analysed by MS18. Although neither GFC, LC nor MS are new technologies, coupling these separation and detection technologies with microfluidic systems has enabled the development of HTS capabilities.

With recent advances in nuclear magnetic resonance (NMR) spectroscopy, a number of NMR-based binding methods have been developed to monitor NMR signal changes from the receptor or a ligand upon receptor-ligand binding. With sequence-specific resonance assignments, the methods provide direct information on the binding affinity of the screening compounds as well as the binding location on soluble receptors19,20. NMR screening enables the detection of very weakly bound ligands (Kd ~10 mM) and as such is especially suited to Lead Generation applications where the binding orientation of very small compounds (i.e. drug fragments) can be exploited to rapidly deliver new lead series.

More label-free assays have been reviewed recently21,22. Though many label-free methods do not have high throughput for compound screening, they do facilitate hit validation and lead optimisation for soluble receptor targets. Membrane-bound receptors, however, still present challenges for these label-free technologies.

In summary, radioligand binding assays continue to represent a critical capability in receptor research. Heterologous expression systems have enabled the production of relatively high concentrations of receptor protein which, when combined with homogeneous binding formats and automated liquid handling systems, has enabled a step-jump improvement in the technical ease and throughput of binding assays. As detection technologies advance, new receptor binding methods emerge that provide detailed information on receptor-ligand interactions that were previously not possible. Their appropriate application in drug discovery may further enable the identification and optimisation of new chemical entities to treat human disease.

Acknowledgement

Thanks to Dr. Deborah.S.Hartman for her critical review of this article.

References

1. Lefkowitz, R.J., Roth, J. Pricer, W., and Pastan, I. ACTH receptors in the adrenal: Specific binding of ACTH-1251 and its relation to adenyl cyclase. Proc. Natl. Acad. Sci. U.S.A. 65, 745–752 (1970)
2. Bylund DB, Deupree JD, Toews ML. Radioligand-binding methods for membrane preparations and intact cells. Methods Mol Biol., 259,1-28, (2004)
3. Liu J, Zacco A, Piser TM, Scott CW Microplate gel-filtration method for radioligand-binding assays. Anal Biochem. 308, 127-33 (2002)
4. Hart HE, Greenwald EB. Scintillation proximity assay (SPA)—a new method of immunoassay. Direct and inhibition mode detection with human albumin and rabbit antihuman albumin. Mol Immunol. 16, 265-7 (1979)
5. Bryant R, McGuinness D, Turek-Etienne T, Guyer D, Yu L, Howells L, Caravano J, Zhai Y, Lachowicz J. WGA-coated yttrium oxide beads enable an imaging-based adenosine 2a receptor binding scintillation proximity assay suitable for high throughput screening. Assay Drug Dev Technol. 2, 290-9 (2004)
6. Ohno K, Fukushima T, Santa T, Waizumi N, Tokuyama H, Maeda M, Imai K. Estrogen receptor binding assay method for endocrine disruptors using fluorescence polarization. Anal Chem. 74, 4391-6 (2002)
7. Lee PH, Bevis DJ. Development of a homogeneous high throughput fluorescence polarization assay for G protein-coupled receptor binding. J Biomol Screen. 5, 415-19 (2000)
8. Lo MC, Aulabaugh A, Jin G, Cowling R, Bard J, Malamas M, Ellestad G. Evaluation of fluorescence-based thermal shift assays for hit identification in drug discovery. Anal Biochem. 332, 153-9 (2004)
9. Pantoliano MW, Petrella EC, Kwasnoski JD, Lobanov VS, Myslik J, Graf E, Carver T, Asel E, Springer BA, Lane P, Salemme FR. High-density miniaturized thermal shift assays as a general strategy for drug discovery. J Biomol Screen. 6, 429-40 (2001)
10. Karger BL, Cohen AS, Guttman A. High-performance capillary electrophoresis in the biological sciences. J Chromatogr. 492, 585-614 (1989)
11. Pierceall WE, Zhang L, Hughes DE. Affinity capillary electrophoresis analyses of protein-protein interactions in target-directed drug discovery. Methods Mol Biol.261, 187-98 (2004)
12. Moore KJ, Turconi S, Ashman S, Ruediger M, Haupts U, Emerick V V, Pope AJ. Single Molecule Detection Technologies in Miniaturized High Throughput Screening: Fluorescence Correlation Spectroscopy. J Biomol Screen. 4, 335-354 (1999)
13. Scheel AA, Funsch B, Busch M, Gradl G, Pschorr J, Lohse MJ. Receptor-ligand interactions studied with homogeneous fluorescence-based assays suitable for miniaturized screening. J Biomol Screen. 6, 11-8 (2001)
14. Rudiger M, Haupts U, Moore KJ, Pope AJ. Single-molecule detection technologies in miniaturized high throughput screening: binding assays for g protein-coupled receptors using fluorescence intensity distribution analysis and fluorescence anisotropy. J Biomol Screen. 6, 29-37 (2001)
15. Pramanik A. Ligand-receptor interactions in live cells by fluorescence correlation spectroscopy. Curr Pharm Biotechnol. 5, 205-12 (2004); Jager S, Brand L, Eggeling C. New fluorescence techniques for high-throughput drug discovery. Curr Pharm Biotechnol. 4, 463-76 (2003)
16. Cooper MA. Label-free screening of bio-molecular interactions Anal Bioanal Chem. 377, 834-42 (2003)
17. Lofas S. Optimizing the hit-to-lead process using SPR analysis. Assay Drug Dev Technol. 2, 407-15 (2004)
18. Annis DA, Nazef N, Chuang CC, Scott MP, Nash HM. A general technique to rank protein-ligand binding affinities and determine allosteric versus direct binding site competition in compound mixtures. J Am Chem Soc. 126, 15495-503 (2004)
19. Vogtherr M, Fiebig K. NMR-based screening methods for lead discovery. EXS. 93, 183-202 (2003)
20. Zartler ER, Yan J, Mo H, Kline AD, Shapiro MJ. ID NMR Methods in ligand-receptor interactions. Curr Top Med Chem. 3, 25-37 (2003)
21. Cooper MA. Advances in membrane receptor screening and analysis. J Mol Recognition. 17, 286-315 (2004)
22. Comess KM, Schurdak ME. Affinity-based screening techniques for enhancing lead discovery. Curr Opin Drug Discov Devel. 7, 411-6 (2004)